[ { "path": ".babelrc", "content": "{\n \"presets\": [\n [\n \"@babel/preset-env\",\n {\n \"modules\": false\n }\n ],\n \"@babel/preset-react\"\n ],\n \"plugins\": [\n [\n \"@babel/plugin-proposal-decorators\",\n {\n \"legacy\": true\n }\n ],\n \"@babel/plugin-proposal-class-properties\",\n \"@babel/plugin-syntax-dynamic-import\",\n \"@babel/plugin-syntax-import-meta\",\n \"@babel/plugin-proposal-json-strings\",\n \"@babel/plugin-proposal-function-sent\",\n \"@babel/plugin-proposal-export-namespace-from\",\n \"@babel/plugin-proposal-numeric-separator\",\n \"@babel/plugin-proposal-throw-expressions\",\n \"@babel/plugin-proposal-optional-chaining\"\n ],\n \"env\": {\n \"test\": {\n \"presets\": [\n [\n \"@babel/preset-env\",\n {\n \"modules\": \"auto\"\n }\n ],\n \"@babel/preset-react\"\n ]\n }\n }\n}\n" }, { "path": ".gitignore", "content": "### PYTHON\n#\n# Byte-compiled / optimized / DLL files\n__pycache__/\n*.py[cod]\n*$py.class\n\n# C extensions\n*.so\n\n# Distribution / packaging\n.Python\nbuild/\ndevelop-eggs/\ndist/\ndownloads/\neggs/\n.eggs/\nlib/\nlib64/\nparts/\nsdist/\nvar/\nwheels/\n*.egg-info/\n.installed.cfg\n*.egg\nMANIFEST\n\n# PyInstaller\n# Usually these files are written by a python script from a template\n# before PyInstaller builds the exe, so as to inject date/other infos into it.\n*.manifest\n*.spec\n\n# Installer logs\npip-log.txt\npip-delete-this-directory.txt\n\n# Unit test / coverage reports\nhtmlcov/\n.tox/\n.coverage\n.coverage.*\n.cache\nnosetests.xml\ncoverage.xml\n*.cover\n.hypothesis/\n\n# Translations\n*.mo\n*.pot\n\n# Django stuff:\n*.log\n.static_storage/\n.media/\nlocal_settings.py\n\n# Flask stuff:\ninstance/\n.webassets-cache\n\n# Scrapy stuff:\n.scrapy\n\n# Sphinx documentation\ndocs/_build/\n\n# PyBuilder\ntarget/\n\n# Jupyter Notebook\n.ipynb_checkpoints\n\n# pyenv\n.python-version\n\n# celery beat schedule file\ncelerybeat-schedule\n\n# SageMath parsed files\n*.sage.py\n\n# Environments\n.env\n.venv\nenv/\nvenv/\nENV/\nenv.bak/\nvenv.bak/\n\n# Spyder project settings\n.spyderproject\n.spyproject\n\n# Rope project settings\n.ropeproject\n\n# mkdocs documentation\n/site\n\n# mypy\n.mypy_cache/\n\n\n### NODE\n# Logs\nlogs\n*.log\nnpm-debug.log*\nyarn-debug.log*\nyarn-error.log*\n\n# Runtime data\npids\n*.pid\n*.seed\n*.pid.lock\n\n# Directory for instrumented libs generated by jscoverage/JSCover\nlib-cov\n\n# Coverage directory used by tools like istanbul\ncoverage\n\n# nyc test coverage\n.nyc_output\n\n# Grunt intermediate storage (http://gruntjs.com/creating-plugins#storing-task-files)\n.grunt\n\n# Bower dependency directory (https://bower.io/)\nbower_components\n\n# node-waf configuration\n.lock-wscript\n\n# Compiled binary addons (https://nodejs.org/api/addons.html)\nbuild/Release\n\n# Dependency directories\nnode_modules/\njspm_packages/\n\n# Typescript v1 declaration files\ntypings/\n\n# Optional npm cache directory\n.npm\n\n# Optional eslint cache\n.eslintcache\n\n# Optional REPL history\n.node_repl_history\n\n# Output of 'npm pack'\n*.tgz\n\n# Yarn Integrity file\n.yarn-integrity\n\n# dotenv environment variables file\n.env\n\n# next.js build output\n.next\n\n# vim custom added\n*.swp\n*.swo\n\n" }, { "path": ".prettierignore", "content": "package-lock.json\npackage.json\n" }, { "path": "LICENSE", "content": "The MIT License (MIT)\n\nCopyright (c) 2017-2018 Alexander Putilin\n\nPermission is hereby granted, free of charge, to any person obtaining a copy\nof this software and associated documentation files (the \"Software\"), to deal\nin the Software without restriction, including without limitation the rights\nto use, copy, modify, merge, publish, distribute, sublicense, and/or sell\ncopies of the Software, and to permit persons to whom the Software is\nfurnished to do so, subject to the following conditions:\n\nThe above copyright notice and this permission notice shall be included in all\ncopies or substantial portions of the Software.\n\nTHE SOFTWARE IS PROVIDED \"AS IS\", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR\nIMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,\nFITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE\nAUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER\nLIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,\nOUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE\nSOFTWARE.\n" }, { "path": "README.md", "content": "# Inside Python Dict - an explorable explanation\n\nThis repository contains the code for [\"Inside Python Dict\"](https://just-taking-a-ride.com/inside_python_dict/), a explorable explanation of python dicts. \n\n## Code\n\nThe code is quite messy, the build system works on my laptop but might not work on your machine. That's all fixable, but I am not sure if there is any interest from people in using the actual code. If so, let me know, I'll clean things up a bit and write an overview of the codebase. \n\nMeanwhile, try running `npm install && npm start` and see if if works. \n\nBugfixes will be gladly accepted.\n\n### Disclaimer\n\nI am providing code in the repository to you under an open source license. Because this is my personal repository, the license you receive to my code is from me and not my employer (Facebook)\n" }, { "path": "package.json", "content": "{\n \"name\": \"inside_python_dict\",\n \"version\": \"1.0.0\",\n \"description\": \"Inside python dict - an explorable explanation\",\n \"main\": \"index.js\",\n \"scripts\": {\n \"jest\": \"jest --env=node\",\n \"test:pystress\": \"npm run extractcode && ./stress_test_python.sh\",\n \"test:pyunit\": \"npm run extractcode && ./unittest_python.sh\",\n \"test\": \"npm run jest && npm run test:pyunit && npm run test:pystress\",\n \"build:ssr\": \"./ssr-all.sh\",\n \"update:html\": \"mkdir -p build && (for i in {chapter1,chapter2,chapter3,chapter4}; do mustache src/mustache/$i.json src/page.html.template > src/autogenerated/$i.html; done)\",\n \"start\": \"webpack-dev-server --config webpack.dev.js --host 0.0.0.0\",\n \"serve\": \"http-server -p 9090 dist/\",\n \"build\": \"npm run build:ssr && webpack --config webpack.prod.js\",\n \"babel-node\": \"npx babel-node --presets '@babel/env'\",\n \"extractcode\": \"mkdir -p build && npm run babel-node scripts/extractPythonCode.js\",\n \"dictserver\": \"rm pynode.sock ; npm run babel-node scripts/pyReimplWrapper.js; rm pynode.sock\",\n \"postinstall\": \"patch-package\"\n },\n \"repository\": {\n \"type\": \"git\",\n \"url\": \"git+https://github.com/eleweek/inside_python_dict.git\"\n },\n \"author\": \"Alexander Putilin\",\n \"license\": \"MIT\",\n \"bugs\": {\n \"url\": \"https://github.com/eleweek/inside_python_dict/issues\"\n },\n \"homepage\": \"https://github.com/eleweek/inside_python_dict#readme\",\n \"prettier\": {\n \"printWidth\": 120,\n \"tabWidth\": 4,\n \"useTabs\": false,\n \"singleQuote\": true,\n \"bracketSpacing\": false,\n \"semi\": true,\n \"trailingComma\": \"es5\"\n },\n \"dependencies\": {\n \"@fortawesome/fontawesome-svg-core\": \"^1.2.10\",\n \"@fortawesome/free-brands-svg-icons\": \"^5.6.1\",\n \"@fortawesome/free-solid-svg-icons\": \"^5.6.1\",\n \"@fortawesome/react-fontawesome\": \"^0.1.3\",\n \"bignumber.js\": \"^8.0.1\",\n \"bootstrap\": \"^4.1.3\",\n \"bowser\": \"^2.0.0-beta.3\",\n \"classnames\": \"^2.2.6\",\n \"d3\": \"^5.7.0\",\n \"d3-selection-multi\": \"^1.0.1\",\n \"i\": \"^0.3.6\",\n \"immutable\": \"^4.0.0-rc.12\",\n \"lodash\": \"^4.17.11\",\n \"lowlight\": \"^1.11.0\",\n \"memoize-one\": \"^4.1.0\",\n \"mobx\": \"^5.8.0\",\n \"mobx-react\": \"^5.4.3\",\n \"rc-slider\": \"^8.6.4\",\n \"react\": \"^16.6.3\",\n \"react-css-transition-replace\": \"^3.0.3\",\n \"react-dom\": \"^16.6.3\",\n \"react-error-boundary\": \"^1.2.3\",\n \"react-input-autosize\": \"^2.2.1\",\n \"react-popper\": \"^1.3.2\",\n \"react-smooth-scrollbar\": \"^8.0.6\",\n \"react-stickynode\": \"^2.1.0\",\n \"rehype\": \"^7.0.0\",\n \"smooth-scrollbar\": \"8.3.1\"\n },\n \"devDependencies\": {\n \"@babel/cli\": \"^7.2.0\",\n \"@babel/core\": \"^7.2.2\",\n \"@babel/node\": \"^7.2.2\",\n \"@babel/plugin-proposal-class-properties\": \"^7.2.1\",\n \"@babel/plugin-proposal-decorators\": \"^7.2.2\",\n \"@babel/plugin-proposal-export-namespace-from\": \"^7.2.0\",\n \"@babel/plugin-proposal-function-sent\": \"^7.2.0\",\n \"@babel/plugin-proposal-json-strings\": \"^7.2.0\",\n \"@babel/plugin-proposal-numeric-separator\": \"^7.2.0\",\n \"@babel/plugin-proposal-object-rest-spread\": \"^7.2.0\",\n \"@babel/plugin-proposal-optional-chaining\": \"^7.2.0\",\n \"@babel/plugin-proposal-throw-expressions\": \"^7.2.0\",\n \"@babel/plugin-syntax-dynamic-import\": \"^7.2.0\",\n \"@babel/plugin-syntax-import-meta\": \"^7.2.0\",\n \"@babel/plugin-transform-destructuring\": \"^7.2.0\",\n \"@babel/preset-env\": \"^7.2.0\",\n \"@babel/preset-react\": \"^7.0.0\",\n \"babel-core\": \"^7.0.0-bridge.0\",\n \"babel-jest\": \"^23.4.2\",\n \"babel-loader\": \"^8.0.0\",\n \"babel-plugin-lodash\": \"^3.3.4\",\n \"clean-webpack-plugin\": \"^1.0.0\",\n \"css-loader\": \"^2.0.1\",\n \"dotenv\": \"^6.2.0\",\n \"html-webpack-plugin\": \"^3.2.0\",\n \"http-server\": \"^0.11.1\",\n \"husky\": \"^1.2.1\",\n \"ignore-styles\": \"^5.0.1\",\n \"jest\": \"^23.6.0\",\n \"mini-css-extract-plugin\": \"^0.5.0\",\n \"mustache\": \"^3.0.1\",\n \"npm\": \"^6.5.0\",\n \"patch-package\": \"^5.1.1\",\n \"prettier\": \"^1.15.3\",\n \"pretty-quick\": \"^1.8.0\",\n \"split\": \"^1.0.1\",\n \"style-loader\": \"^0.23.1\",\n \"unminified-webpack-plugin\": \"^2.0.0\",\n \"webpack\": \"^4.27.1\",\n \"webpack-bundle-analyzer\": \"^3.0.3\",\n \"webpack-cli\": \"^3.1.2\",\n \"webpack-dev-server\": \"^3.1.10\",\n \"webpack-merge\": \"^4.1.5\"\n },\n \"husky\": {\n \"hooks\": {\n \"pre-commit\": \"pretty-quick --staged\"\n }\n }\n}\n" }, { "path": "patches/python32_debug.diff", "content": "diff --git a/Objects/dictobject.c b/Objects/dictobject.c\nindex c10bfccdce..3734a08281 100644\n--- a/Objects/dictobject.c\n+++ b/Objects/dictobject.c\n@@ -321,6 +321,8 @@ lookdict(PyDictObject *mp, PyObject *key, register Py_hash_t hash)\n PyObject *startkey;\n \n i = (size_t)hash & mask;\n+ fprintf(stderr, \"lookdict hash = %ld\\n\", hash);\n+ fprintf(stderr, \"initial i = %zu\\n\", i);\n ep = &ep0[i];\n if (ep->me_key == NULL || ep->me_key == key)\n return ep;\n@@ -355,7 +357,9 @@ lookdict(PyDictObject *mp, PyObject *key, register Py_hash_t hash)\n least likely outcome, so test for that last. */\n for (perturb = hash; ; perturb >>= PERTURB_SHIFT) {\n i = (i << 2) + i + perturb + 1;\n+ fprintf(stderr, \"next i = %zu perturb = %zu\\n\", i, perturb);\n ep = &ep0[i & mask];\n+ fprintf(stderr, \"next i & mask = %zu perturb = %zu\\n\", i & mask, perturb);\n if (ep->me_key == NULL)\n return freeslot == NULL ? ep : freeslot;\n if (ep->me_key == key)\n@@ -648,6 +652,7 @@ dictresize(PyDictObject *mp, Py_ssize_t minused)\n }\n }\n else {\n+ fprintf(stderr, \"PyMem_NEW branch\");\n newtable = PyMem_NEW(PyDictEntry, newsize);\n if (newtable == NULL) {\n PyErr_NoMemory();\n@@ -693,6 +698,7 @@ PyObject *\n _PyDict_NewPresized(Py_ssize_t minused)\n {\n PyObject *op = PyDict_New();\n+ fprintf(stderr, \"_PyDict_NewPresized() %p %d\\n\", op, (int)minused);\n \n if (minused>5 && op != NULL && dictresize((PyDictObject *)op, minused) == -1) {\n Py_DECREF(op);\ndiff --git a/Objects/longobject.c b/Objects/longobject.c\nindex e2a4ef9c5e..7d72c88417 100644\n--- a/Objects/longobject.c\n+++ b/Objects/longobject.c\n@@ -2611,6 +2611,7 @@ long_hash(PyLongObject *v)\n sign = -1;\n i = -(i);\n }\n+ fprintf(stderr, \"i = %ld\\n\", i);\n while (--i >= 0) {\n /* Here x is a quantity in the range [0, _PyHASH_MODULUS); we\n want to compute x * 2**PyLong_SHIFT + v->ob_digit[i] modulo\n" }, { "path": "patches/smooth-scrollbar+8.3.1.patch", "content": "patch-package\n--- a/node_modules/smooth-scrollbar/events/touch.js\n+++ b/node_modules/smooth-scrollbar/events/touch.js\n@@ -1,7 +1,7 @@\n import { eventScope, TouchRecord, } from '../utils/';\n var activeScrollbar;\n export function touchHandler(scrollbar) {\n- var MIN_EAING_MOMENTUM = 50;\n+ var MIN_EAING_MOMENTUM = 3;\n var EASING_MULTIPLIER = /Android/.test(navigator.userAgent) ? 3 : 2;\n var target = scrollbar.options.delegateTo || scrollbar.containerEl;\n var touchRecord = new TouchRecord();\n--- a/node_modules/smooth-scrollbar/geometry/update.js\n+++ b/node_modules/smooth-scrollbar/geometry/update.js\n@@ -4,6 +4,9 @@ export function update(scrollbar) {\n x: Math.max(newSize.content.width - newSize.container.width, 0),\n y: Math.max(newSize.content.height - newSize.container.height, 0),\n };\n+ // hack for a weird chrome on windows bug\n+ if (limit.x <= 2) limit.x = 0;\n+ if (limit.y <= 2) limit.y = 0;\n // metrics\n var containerBounding = scrollbar.containerEl.getBoundingClientRect();\n var bounding = {\n--- a/node_modules/smooth-scrollbar/scrollbar.js\n+++ b/node_modules/smooth-scrollbar/scrollbar.js\n@@ -322,6 +322,10 @@ var Scrollbar = /** @class */ (function () {\n if (limit.x === 0 && limit.y === 0) {\n this._updateDebounced();\n }\n+ if (Math.abs(deltaY) > Math.abs(deltaX)) {\n+ if (deltaY > 0 && offset.y === limit.y) return true;\n+ if (deltaY < 0 && offset.y === 0) return true;\n+ }\n var destX = clamp(deltaX + offset.x, 0, limit.x);\n var destY = clamp(deltaY + offset.y, 0, limit.y);\n var res = true;\n--- a/node_modules/smooth-scrollbar/track/track.js\n+++ b/node_modules/smooth-scrollbar/track/track.js\n@@ -41,8 +41,9 @@ var ScrollbarTrack = /** @class */ (function () {\n this.element.classList.remove('show');\n };\n ScrollbarTrack.prototype.update = function (scrollOffset, containerSize, pageSize) {\n+ // -2 is a hack for a weird chrome on windows bug\n setStyle(this.element, {\n- display: pageSize <= containerSize ? 'none' : 'block',\n+ display: pageSize - 2 <= containerSize ? 'none' : 'block',\n });\n this.thumb.update(scrollOffset, containerSize, pageSize);\n };\ndeleted file mode 100644\n--- a/node_modules/smooth-scrollbar/track/track.js.map\n+++ /dev/null\n@@ -1 +0,0 @@\n-{\"version\":3,\"file\":\"track.js\",\"sourceRoot\":\"\",\"sources\":[\"../src/track/track.ts\"],\"names\":[],\"mappings\":\"AAEA,OAAO,EAAE,cAAc,EAAE,MAAM,SAAS,CAAC;AAEzC,OAAO,EACL,QAAQ,GACT,MAAM,WAAW,CAAC;AAEnB;IAUE,wBACE,SAAyB,EACzB,YAAwB;QAAxB,6BAAA,EAAA,gBAAwB;QAT1B;;WAEG;QACM,YAAO,GAAG,QAAQ,CAAC,aAAa,CAAC,KAAK,CAAC,CAAC;QAEzC,aAAQ,GAAG,KAAK,CAAC;QAMvB,IAAI,CAAC,OAAO,CAAC,SAAS,GAAG,qCAAmC,SAAW,CAAC;QAExE,IAAI,CAAC,KAAK,GAAG,IAAI,cAAc,CAC7B,SAAS,EACT,YAAY,CACb,CAAC;QAEF,IAAI,CAAC,KAAK,CAAC,QAAQ,CAAC,IAAI,CAAC,OAAO,CAAC,CAAC;IACpC,CAAC;IAED;;;;OAIG;IACH,iCAAQ,GAAR,UAAS,kBAA+B;QACtC,kBAAkB,CAAC,WAAW,CAAC,IAAI,CAAC,OAAO,CAAC,CAAC;IAC/C,CAAC;IAED;;OAEG;IACH,6BAAI,GAAJ;QACE,EAAE,CAAC,CAAC,IAAI,CAAC,QAAQ,CAAC,CAAC,CAAC;YAClB,MAAM,CAAC;QACT,CAAC;QAED,IAAI,CAAC,QAAQ,GAAG,IAAI,CAAC;QACrB,IAAI,CAAC,OAAO,CAAC,SAAS,CAAC,GAAG,CAAC,MAAM,CAAC,CAAC;IACrC,CAAC;IAED;;OAEG;IACH,6BAAI,GAAJ;QACE,EAAE,CAAC,CAAC,CAAC,IAAI,CAAC,QAAQ,CAAC,CAAC,CAAC;YACnB,MAAM,CAAC;QACT,CAAC;QAED,IAAI,CAAC,QAAQ,GAAG,KAAK,CAAC;QACtB,IAAI,CAAC,OAAO,CAAC,SAAS,CAAC,MAAM,CAAC,MAAM,CAAC,CAAC;IACxC,CAAC;IAED,+BAAM,GAAN,UACE,YAAoB,EACpB,aAAqB,EACrB,QAAgB;QAEhB,QAAQ,CAAC,IAAI,CAAC,OAAO,EAAE;YACrB,OAAO,EAAE,QAAQ,IAAI,aAAa,CAAC,CAAC,CAAC,MAAM,CAAC,CAAC,CAAC,OAAO;SACtD,CAAC,CAAC;QAEH,IAAI,CAAC,KAAK,CAAC,MAAM,CAAC,YAAY,EAAE,aAAa,EAAE,QAAQ,CAAC,CAAC;IAC3D,CAAC;IACH,qBAAC;AAAD,CAAC,AApED,IAoEC\"}\n\\ No newline at end of file\n" }, { "path": "patches/subscribe-ui-event+2.0.4.patch", "content": "patch-package\nnew file mode 100644\nBinary files /dev/null and b/node_modules/subscribe-ui-event/.index.es.js.swp differ\n--- a/node_modules/subscribe-ui-event/index.es.js\n+++ b/node_modules/subscribe-ui-event/index.es.js\n@@ -7,7 +7,7 @@ import listenLib from './dist-es/lib/listen';\n import subscribeLib from './dist-es/subscribe';\n import unsubscribeLib from './dist-es/unsubscribe';\n \n-const IS_CLIENT = typeof window !== 'undefined';\n+var IS_CLIENT = typeof window !== 'undefined';\n \n function warn() {\n if (process.env.NODE_ENV !== 'production') {\n@@ -15,6 +15,6 @@ function warn() {\n }\n }\n \n-export const listen = IS_CLIENT ? listenLib : warn;\n-export const subscribe = IS_CLIENT ? subscribeLib : warn;\n-export const unsubscribe = IS_CLIENT ? unsubscribeLib : warn;\n+export var listen = IS_CLIENT ? listenLib : warn;\n+export var subscribe = IS_CLIENT ? subscribeLib : warn;\n+export var unsubscribe = IS_CLIENT ? unsubscribeLib : warn;\n" }, { "path": "python_code/actual_dict_factory_test.py", "content": "import unittest\n\nfrom dict32_reimplementation_test_v2 import dict_factory\nfrom dictinfo import dump_py_dict\n\n\ndef table_size(d):\n return len(dump_py_dict(d)[0])\n\n\nclass TestDictFactory(unittest.TestCase):\n def test_dict_factory(self):\n self.assertEqual(table_size(dict_factory([])), 8)\n self.assertEqual(table_size(dict_factory([(1, 1)])), 8)\n self.assertEqual(table_size(dict_factory([(1, 1), (1, 2), (1, 3), (1, 4)])), 8)\n self.assertEqual(table_size(dict_factory([(1, 1), (1, 2), (1, 3), (1, 4), (1, 5)])), 8)\n self.assertEqual(table_size(dict_factory([(1, 1), (1, 2), (1, 3), (1, 4), (1, 5), (1, 6), (1, 7), (1, 8)])), 16)\n self.assertEqual(table_size({1: 1, 1: 2, 1: 3, 1: 4, 1: 5, 1: 6, 1: 7, 1: 8}), 16)\n self.assertEqual(table_size(dict([(1, 1), (1, 2), (1, 3), (1, 4), (1, 5), (1, 6), (1, 7), (1, 8)])), 8)\n\n self.assertEqual(table_size({\"x\": \"y\", \"abde\": 1, \"cdef\": 4, \"world\": 9, \"hmmm\": 16, \"hello\": 25, \"xxx\": 36, \"ya\": 49, \"hello,world!\": 64, \"well\": 81, \"meh\": 100}), 64)\n\n\ndef main():\n unittest.main()\n\n\nif __name__ == \"__main__\":\n main()\n" }, { "path": "python_code/build_autogenerated_chapter1_hash.py", "content": "import os\nimport sys\nsys.path.append(os.path.join(os.path.dirname(os.path.abspath(__file__)), '..', 'build'))\nfrom hash_chapter1_extracted import *\n\n\ndef create_new(numbers):\n return build_insert_all(numbers)\n\n\ndef create_new_broken(numbers):\n return build_not_quite_what_we_want(numbers)\n\n\ndef has_key(keys, key):\n return has_number(keys, key)\n\n\ndef linear_search(numbers, number):\n return simple_search(numbers, number)\n" }, { "path": "python_code/build_autogenerated_chapter2.py", "content": "import os\nimport sys\nsys.path.append(os.path.join(os.path.dirname(os.path.abspath(__file__)), '..', 'build'))\nfrom hash_chapter2_extracted import *\n" }, { "path": "python_code/build_autogenerated_chapter3_chapter4.py", "content": "import os\nimport sys\nsys.path.append(os.path.join(os.path.dirname(os.path.abspath(__file__)), '..', 'build'))\nfrom dict32js_extracted import Dict32Extracted\nfrom hash_class_recycling_extracted import HashClassRecyclingExtracted\nfrom hash_class_no_recycling_extracted import HashClassNoRecyclingExtracted\n" }, { "path": "python_code/chapter1_linear_search_reimplementation_test.py", "content": "import random\nimport argparse\n\nimport hash_chapter1_reimpl_js\nimport hash_chapter1_impl\nimport build_autogenerated_chapter1_hash\n\nIMPLEMENTATIONS = {\n 'ref': hash_chapter1_impl.linear_search,\n 'js': hash_chapter1_reimpl_js.linear_search,\n 'py_extracted': build_autogenerated_chapter1_hash.linear_search\n}\n\n\ndef run(test_implementation, size):\n MAX_VAL = 5000\n ref_search = IMPLEMENTATIONS['ref']\n test_search = IMPLEMENTATIONS[test_implementation]\n\n numbers = [random.randint(-MAX_VAL, MAX_VAL) for _ in range(size)]\n\n for number in numbers:\n assert ref_search(numbers, number)\n assert test_search(numbers, number)\n\n for i in range(size * 3):\n number = random.randint(-MAX_VAL, MAX_VAL)\n assert ref_search(numbers, number) == test_search(numbers, number)\n\n\nif __name__ == \"__main__\":\n parser = argparse.ArgumentParser(description='Stress-test chapter1 reimplementation')\n parser.add_argument('--test-implementation', choices=['py_extracted', 'js'], required=True)\n parser.add_argument('--size', type=int, default=100)\n args = parser.parse_args()\n\n run(test_implementation=args.test_implementation,\n size=args.size)\n" }, { "path": "python_code/chapter4_probing_python_reimplementation_test.py", "content": "import json\nfrom common import AllKeyValueFactory\nfrom js_reimpl_common import _init_sock_stuff, dump_simple_py_obj\nfrom pprint import pprint\n\nsock, sockfile = _init_sock_stuff()\n\n\ndef probe_all_js(key, slots_count):\n global sockfile\n global sock\n\n data = {\n \"dict\": \"pythonProbing\",\n \"args\": {\n 'key': dump_simple_py_obj(key),\n 'slotsCount': slots_count\n },\n }\n\n sock.send(bytes(json.dumps(data) + \"\\n\", 'UTF-8'))\n response = json.loads(sockfile.readline())\n\n return response['result']\n\n\ndef probe_all(key, slots_count=8):\n PERTURB_SHIFT = 5\n links = [[] for _ in range(slots_count)]\n hash_code = hash(key)\n perturb = 2**64 + hash_code if hash_code < 0 else hash_code\n idx = hash_code % slots_count\n start_idx = idx\n visited = set()\n while len(visited) < slots_count:\n visited.add(idx)\n next_idx = (idx * 5 + perturb + 1) % slots_count\n links[idx].append({'nextIdx': next_idx, 'perturbLink': perturb != 0})\n idx = next_idx\n perturb >>= PERTURB_SHIFT\n\n return {'startIdx': start_idx, 'links': links}\n\n\ndef test():\n factory = AllKeyValueFactory(100)\n for slots_count in [8, 16, 32]:\n for i in range(300):\n key = factory.generate_key()\n assert probe_all(key, slots_count) == probe_all_js(key, slots_count)\n\n\nif __name__ == \"__main__\":\n test()\n" }, { "path": "python_code/common.py", "content": "import random\nimport string\n\n\nclass EmptyValueClass(object):\n def __str__(self):\n return \"EMPTY\"\n\n def __repr__(self):\n return \"\"\n\n\nclass DummyValueClass(object):\n def __str__(self):\n return \"\"\n\n def __repr__(self):\n return \"\"\n\n\nEMPTY = EmptyValueClass()\nDUMMY = DummyValueClass()\n\n\ndef get_object_field_or_null(obj, field_name):\n try:\n return getattr(obj, field_name)\n except ValueError:\n return EMPTY\n\n\ndef get_object_field_or_none(obj, field_name):\n try:\n return getattr(obj, field_name)\n except ValueError:\n return None\n\n\ndef generate_random_string(str_len=5):\n # FROM: https://stackoverflow.com/questions/2257441/random-string-generation-with-upper-case-letters-and-digits-in-python\n return ''.join(random.choice(string.ascii_uppercase + string.digits) for _ in range(str_len))\n\n\n_unicode_chars = string.ascii_uppercase + string.digits + \"йцукенгшщзхъфывапролджэячсмитьбю\"\n\n\ndef generate_random_unicode(str_len):\n # FROM: https://stackoverflow.com/questions/2257441/random-string-generation-with-upper-case-letters-and-digits-in-python\n return ''.join(random.choice(_unicode_chars) for _ in range(str_len))\n\n\nclass IntKeyValueFactory(object):\n def __init__(self, n_inserts):\n self.n_inserts = n_inserts\n self._insert_count = 0\n self._key_range = list(range(n_inserts))\n\n def generate_key(self):\n return random.choice(self._key_range)\n\n def generate_value(self):\n self._insert_count += 1\n return self._insert_count\n\n\nclass AllKeyValueFactory(object):\n def __init__(self, n_inserts, int_chance=0.1, long_chance=0.1, len0_chance=0.01, len1_chance=0.1, len2_chance=0.3, len3_chance=0.2, len_random_chance=0.17):\n self.int_pbf = int_chance\n self.long_pbf = self.int_pbf + long_chance\n self.len0_pbf = self.int_pbf + len0_chance\n self.len1_pbf = self.len0_pbf + len1_chance\n self.len2_pbf = self.len1_pbf + len2_chance\n self.len3_pbf = self.len2_pbf + len3_chance\n self.len_random_pbf = self.len3_pbf + len_random_chance\n assert 0.0 <= self.len3_pbf <= 1.0\n\n half_range = int(n_inserts / 2)\n self._int_range = [i - half_range for i in range(2 * half_range)]\n\n def _generate_obj(self):\n r = random.random()\n if r <= self.int_pbf:\n return random.choice(self._int_range)\n if r <= self.long_pbf:\n sign = \"-\" if random.random() < 0.5 else \"\"\n first_digit = random.choice(\"123456789\")\n return sign + first_digit + ''.join(random.choice(\"0123456789\") for _ in range(random.randint(20, 50)))\n if r <= self.len0_pbf:\n return \"\"\n if r <= self.len1_pbf:\n return generate_random_unicode(1)\n if r <= self.len2_pbf:\n return generate_random_unicode(2)\n if r <= self.len3_pbf:\n return generate_random_unicode(3)\n if r <= self.len_random_pbf:\n return generate_random_unicode(random.randint(4, 25))\n return None\n\n def generate_key(self):\n return self._generate_obj()\n\n def generate_value(self):\n return self._generate_obj()\n" }, { "path": "python_code/dict32_reimplementation_test_v2.py", "content": "import random\nimport argparse\nimport json\n\nfrom common import EMPTY, AllKeyValueFactory, IntKeyValueFactory\nfrom dictinfo import dump_py_dict\nfrom dict_reimplementation import PyDictReimplementation32, dump_reimpl_dict\nfrom js_reimplementation_interface import Dict32JsImpl, AlmostPythonDictRecyclingJsImpl, AlmostPythonDictNoRecyclingJsImpl\nimport hash_chapter3_class_impl\nimport build_autogenerated_chapter3_chapter4\n\n\ndef dict_factory(pairs=None):\n if not pairs:\n return {}\n\n # quick&dirty\n def to_string(x):\n return json.dumps(x) if x is not None else \"None\"\n d = eval(\"{\" + \", \".join(\"{}:{}\".format(to_string(k), to_string(v)) for [k, v] in pairs) + \"}\")\n return d\n\n\nIMPLEMENTATIONS = {\n \"dict_actual\": (dict_factory, dump_py_dict),\n \"dict32_reimpl_py\": (PyDictReimplementation32, dump_reimpl_dict),\n \"dict32_reimpl_js\": (Dict32JsImpl, dump_reimpl_dict),\n\n \"dict32_reimpl_py_extracted\": (build_autogenerated_chapter3_chapter4.Dict32Extracted, dump_reimpl_dict),\n\n \"almost_python_dict_recycling_py\": (hash_chapter3_class_impl.AlmostPythonDictImplementationRecycling, dump_reimpl_dict),\n \"almost_python_dict_no_recycling_py\": (hash_chapter3_class_impl.AlmostPythonDictImplementationNoRecycling, dump_reimpl_dict),\n \"almost_python_dict_no_recycling_py_simpler\": (hash_chapter3_class_impl.AlmostPythonDictImplementationNoRecyclingSimplerVersion, dump_reimpl_dict),\n \"almost_python_dict_recycling_js\": (AlmostPythonDictRecyclingJsImpl, dump_reimpl_dict),\n \"almost_python_dict_no_recycling_js\": (AlmostPythonDictNoRecyclingJsImpl, dump_reimpl_dict),\n\n \"almost_python_dict_recycling_py_extracted\": (build_autogenerated_chapter3_chapter4.HashClassRecyclingExtracted, dump_reimpl_dict),\n \"almost_python_dict_no_recycling_py_extracted\": (build_autogenerated_chapter3_chapter4.HashClassNoRecyclingExtracted, dump_reimpl_dict),\n}\n\n\ndef verify_same(d, dump_d_func, dreimpl, dump_dreimpl_func):\n dump_d = dump_d_func(d)\n dump_reimpl = dump_dreimpl_func(dreimpl)\n\n if dump_d != dump_reimpl:\n hashes_orig, keys_orig, values_orig, fill_orig, used_orig = dump_d\n hashes_new, keys_new, values_new, fill_new, used_new = dump_reimpl\n print(\"ORIG SIZE\", len(hashes_orig))\n print(\"NEW SIZE\", len(hashes_new))\n print(\"ORIG fill/used: \", fill_orig, used_orig)\n print(\"NEW fill/used: \", fill_new, used_new)\n if len(hashes_orig) == len(hashes_new):\n size = len(hashes_orig)\n print(\"NEW | ORIG\")\n for i in range(size):\n if hashes_new[i] is not EMPTY or hashes_orig[i] is not EMPTY:\n print(i, \" \" * 3,\n hashes_new[i], keys_new[i], values_new[i], \" \" * 3,\n hashes_orig[i], keys_orig[i], values_orig[i])\n\n assert dump_d == dump_reimpl\n\n\ndef run(ref_impl_factory, ref_impl_dump, test_impl_factory, test_impl_dump, n_inserts, extra_checks, key_value_factory, initial_state, verbose):\n SINGLE_REMOVE_CHANCE = 0.3\n MASS_REMOVE_CHANCE = 0.002\n MASS_REMOVE_COEFF = 0.8\n\n removed = set()\n\n if initial_state:\n d = ref_impl_factory(initial_state)\n else:\n d = ref_impl_factory()\n\n if initial_state:\n dreimpl = test_impl_factory(initial_state)\n else:\n dreimpl = test_impl_factory()\n\n if verbose:\n print(\"Starting test\")\n\n for i in range(n_inserts):\n should_remove = (random.random() < SINGLE_REMOVE_CHANCE)\n if should_remove and d and d.keys(): # TODO: ugly, written while on a plane\n to_remove = random.choice(list(d.keys()))\n if verbose:\n print(\"Removing {}\".format(to_remove))\n del d[to_remove]\n del dreimpl[to_remove]\n if verbose:\n print(d)\n verify_same(d, ref_impl_dump, dreimpl, test_impl_dump)\n removed.add(to_remove)\n\n should_mass_remove = (random.random() < MASS_REMOVE_CHANCE)\n if should_mass_remove and len(d) > 10:\n to_remove_list = random.sample(list(d.keys()), int(MASS_REMOVE_COEFF * len(d)))\n if verbose:\n print(\"Mass-Removing {} elements\".format(len(to_remove_list)))\n for k in to_remove_list:\n del d[k]\n del dreimpl[k]\n removed.add(k)\n\n if extra_checks:\n for k in d.keys():\n assert d[k] == dreimpl[k]\n\n for r in removed:\n try:\n dreimpl[r]\n assert False\n except KeyError:\n pass\n\n key_to_insert = key_value_factory.generate_key()\n value_to_insert = key_value_factory.generate_value()\n _keys_set = getattr(d, '_keys_set', None)\n # TODO: ugly code written on a plane\n # TODO: properly implement in/not in when I land\n if _keys_set is not None:\n key_present = key_to_insert in _keys_set\n else:\n key_present = key_to_insert in d\n\n if not key_present:\n if verbose:\n print(\"Inserting ({key}, {value})\".format(key=key_to_insert, value=value_to_insert))\n try:\n dreimpl[key_to_insert]\n assert False\n except KeyError:\n pass\n else:\n if verbose:\n print(\"Replacing ({key}, {value1}) with ({key}, {value2})\".format(key=key_to_insert, value1=d[key_to_insert], value2=value_to_insert))\n removed.discard(key_to_insert)\n d[key_to_insert] = value_to_insert\n dreimpl[key_to_insert] = value_to_insert\n if verbose:\n print(d)\n verify_same(d, ref_impl_dump, dreimpl, test_impl_dump)\n assert dreimpl[key_to_insert] == value_to_insert\n\n\nif __name__ == \"__main__\":\n parser = argparse.ArgumentParser(description='Stress-test dict-like reimplementations')\n parser.add_argument('--reference-implementation', choices=IMPLEMENTATIONS.keys(), required=True)\n parser.add_argument('--test-implementation', choices=IMPLEMENTATIONS.keys(), required=True)\n parser.add_argument('--no-extra-getitem-checks', dest='extra_checks', action='store_false')\n parser.add_argument('--num-inserts', type=int, default=500)\n parser.add_argument('--forever', action='store_true')\n parser.add_argument('--kv', choices=[\"numbers\", \"all\"], required=True)\n parser.add_argument('--initial-size', type=int, default=-1)\n parser.add_argument('--verbose', action='store_true')\n args = parser.parse_args()\n\n if args.kv == \"numbers\":\n kv_factory = IntKeyValueFactory(args.num_inserts)\n elif args.kv == \"all\":\n kv_factory = AllKeyValueFactory(args.num_inserts)\n\n ref_impl = IMPLEMENTATIONS[args.reference_implementation]\n test_impl = IMPLEMENTATIONS[args.test_implementation]\n\n def test_iteration():\n initial_size = args.initial_size if args.initial_size >= 0 else random.randint(0, 100)\n initial_state = [(kv_factory.generate_key(), kv_factory.generate_value()) for _ in range(initial_size)]\n run(*(ref_impl + test_impl),\n n_inserts=args.num_inserts,\n extra_checks=args.extra_checks,\n key_value_factory=kv_factory,\n initial_state=initial_state,\n verbose=args.verbose)\n\n if args.forever:\n while True:\n test_iteration()\n else:\n test_iteration()\n" }, { "path": "python_code/dict_reimpl_common.py", "content": "from common import EMPTY\n\n\nclass Slot(object):\n def __init__(self, hash_code=EMPTY, key=EMPTY, value=EMPTY):\n self.hash_code = hash_code\n self.key = key\n self.value = value\n\n\nclass BaseDictImpl(object):\n def __init__(self):\n self.slots = [Slot() for _ in range(self.START_SIZE)]\n self.fill = 0\n self.used = 0\n\n def find_nearest_size(self, minused):\n new_size = 8\n while new_size <= minused:\n new_size *= 2\n\n return new_size\n" }, { "path": "python_code/dict_reimplementation.py", "content": "from common import DUMMY, EMPTY\nfrom dict_reimpl_common import BaseDictImpl, Slot\nfrom operator import attrgetter\n\n\nclass PyDictReimplementationBase(BaseDictImpl):\n START_SIZE = 8\n PERTURB_SHIFT = 5\n\n def __init__(self, pairs=None):\n BaseDictImpl.__init__(self)\n start_size = self.find_nearest_size(len(pairs)) if pairs else self.START_SIZE\n self.slots = [Slot() for _ in range(start_size)]\n if pairs:\n for k, v in pairs:\n self[k] = v\n\n def __setitem__(self, key, value):\n hash_code = hash(key)\n perturb = self.signed_to_unsigned(hash_code)\n idx = hash_code % len(self.slots)\n target_idx = None\n while self.slots[idx].key is not EMPTY:\n if self.slots[idx].hash_code == hash_code and self.slots[idx].key == key:\n target_idx = idx\n break\n if target_idx is None and self.slots[idx].key is DUMMY:\n target_idx = idx\n\n idx = (idx * 5 + perturb + 1) % len(self.slots)\n perturb >>= self.PERTURB_SHIFT\n\n if target_idx is None:\n target_idx = idx\n\n if self.slots[target_idx].key is EMPTY:\n self.used += 1\n self.fill += 1\n elif self.slots[target_idx].key is DUMMY:\n self.used += 1\n\n self.slots[target_idx] = Slot(hash_code, key, value)\n if self.fill * 3 >= len(self.slots) * 2:\n self.resize()\n\n def __delitem__(self, key):\n idx = self.lookdict(key)\n\n self.used -= 1\n self.slots[idx].key = DUMMY\n self.slots[idx].value = EMPTY\n\n def __getitem__(self, key):\n idx = self.lookdict(key)\n\n return self.slots[idx].value\n\n @staticmethod\n def signed_to_unsigned(hash_code):\n return 2**64 + hash_code if hash_code < 0 else hash_code\n\n def lookdict(self, key):\n hash_code = hash(key)\n perturb = self.signed_to_unsigned(hash_code)\n\n idx = hash_code % len(self.slots)\n while self.slots[idx].key is not EMPTY:\n if self.slots[idx].hash_code == hash_code and self.slots[idx].key == key:\n return idx\n\n idx = (idx * 5 + perturb + 1) % len(self.slots)\n perturb >>= self.PERTURB_SHIFT\n\n raise KeyError()\n\n def resize(self):\n old_slots = self.slots\n new_size = self.find_nearest_size(self._next_size())\n self.slots = [Slot() for _ in range(new_size)]\n self.fill = self.used\n for slot in old_slots:\n if slot.key is not EMPTY and slot.key is not DUMMY:\n perturb = self.signed_to_unsigned(slot.hash_code)\n idx = slot.hash_code % len(self.slots)\n while self.slots[idx].key is not EMPTY:\n idx = (idx * 5 + perturb + 1) % len(self.slots)\n perturb >>= self.PERTURB_SHIFT\n\n self.slots[idx] = Slot(slot.hash_code, slot.key, slot.value)\n\n\nclass PyDictReimplementation32(PyDictReimplementationBase):\n def _next_size(self):\n return self.used * (4 if self.used <= 50000 else 2)\n\n\ndef dump_reimpl_dict(d):\n def extract_fields(field_name):\n return list(map(attrgetter(field_name), d.slots))\n return extract_fields('hash_code'), extract_fields('key'), extract_fields('value'), d.fill, d.used\n" }, { "path": "python_code/dictinfo.py", "content": "import sys\n\n\ndef dump_py_dict(d):\n vi = sys.version_info\n\n if vi.major != 3:\n raise Exception(\"Unsupported major version\")\n\n if vi.minor < 2:\n raise Exception(\"Unsupported minor version (too old)\")\n if vi.minor > 3:\n raise Exception(\"Unsupported minor version (too new)\")\n\n if vi.minor == 2:\n import dictinfo32\n return dictinfo32.dump_py_dict(d)\n else:\n import dictinfo33\n return dictinfo33.dump_py_dict(d)\n" }, { "path": "python_code/dictinfo32.py", "content": "from ctypes import Structure, c_ulong, POINTER, cast, py_object, c_long\nfrom common import get_object_field_or_null, EMPTY, DUMMY\n\n\nclass PyDictEntry(Structure):\n _fields_ = [\n ('me_hash', c_long),\n ('me_key', py_object),\n ('me_value', py_object),\n ]\n\n\nclass PyDictObject(Structure):\n _fields_ = [\n ('ob_refcnt', c_ulong),\n ('ob_type', c_ulong),\n ('ma_fill', c_ulong),\n ('ma_used', c_ulong),\n ('ma_mask', c_ulong),\n ('ma_table', POINTER(PyDictEntry)),\n ]\n\n\ndef dictobject(d):\n return cast(id(d), POINTER(PyDictObject)).contents\n\n\nd = {0: 0}\ndel d[0]\ndummy_internal = dictobject(d).ma_table[0].me_key\ndel d\n\n\ndef dump_py_dict(d):\n do = dictobject(d)\n\n keys = []\n hashes = []\n values = []\n\n size = do.ma_mask + 1\n\n for i in range(size):\n key = get_object_field_or_null(do.ma_table[i], 'me_key')\n keys.append(key if key is not dummy_internal else DUMMY)\n\n for i, key in enumerate(keys):\n if key is EMPTY:\n hashes.append(EMPTY)\n values.append(EMPTY)\n else:\n hashes.append(do.ma_table[i].me_hash)\n values.append(get_object_field_or_null(do.ma_table[i], 'me_value'))\n\n return hashes, keys, values, do.ma_fill, do.ma_used\n" }, { "path": "python_code/dictinfo33.py", "content": "from ctypes import Structure, c_ulong, POINTER, cast, addressof, py_object, c_long, c_void_p\nfrom common import get_object_field_or_null, EMPTY, DUMMY\n\n\nclass PyDictKeyEntry(Structure):\n _fields_ = [\n ('me_hash', c_long),\n ('me_key', py_object),\n ('me_value', py_object),\n ]\n\n\nclass PyDictKeysObject(Structure):\n _fields_ = [\n ('dk_refcnt', c_long),\n ('dk_size', c_long),\n ('dict_lookup_func', POINTER(c_void_p)),\n ('dk_usable', c_long),\n ('dk_entries', PyDictKeyEntry),\n ]\n\n\nclass PyDictObject(Structure):\n _fields_ = [\n ('ob_refcnt', c_ulong),\n ('ob_type', c_ulong),\n ('ma_used', c_long),\n ('ma_keys', POINTER(PyDictKeysObject)),\n\n # Not actually a void*, split tables are not supported right now\n ('ma_values', POINTER(c_void_p))\n ]\n\n\ndef dictobject(d):\n return cast(id(d), POINTER(PyDictObject)).contents\n\n\nd = {0: 0}\ndel d[0]\ndummy_internal = dictobject(d).ma_keys.contents.dk_entries.me_key\ndel d\n\n\ndef usable_fraction(size):\n return (size * 2 + 1) // 3\n\n\ndef dump_py_dict(d):\n do = dictobject(d)\n\n keys = []\n hashes = []\n values = []\n\n size = do.ma_keys.contents.dk_size\n entries = cast(addressof(do.ma_keys.contents.dk_entries), POINTER(PyDictKeyEntry))\n for i in range(size):\n key = get_object_field_or_null(entries[i], 'me_key')\n keys.append(key if key is not dummy_internal else DUMMY)\n\n for i, key in enumerate(keys):\n if key is EMPTY:\n hashes.append(EMPTY)\n values.append(EMPTY)\n else:\n hashes.append(entries[i].me_hash)\n values.append(get_object_field_or_null(entries[i], 'me_value'))\n\n return hashes, keys, values, usable_fraction(do.ma_keys.contents.dk_size) - do.ma_keys.contents.dk_usable, do.ma_used\n" }, { "path": "python_code/hash_chapter1_impl.py", "content": "def create_new(numbers):\n n = len(numbers)\n keys = [None for i in range(2 * n)]\n\n for num in numbers:\n idx = num % len(keys)\n\n while keys[idx] is not None:\n idx = (idx + 1) % len(keys)\n\n keys[idx] = num\n\n return keys\n\n\ndef create_new_broken(numbers):\n n = len(numbers)\n keys = [None for i in range(n)]\n\n for num in numbers:\n idx = num % len(keys)\n keys[idx] = num\n\n return keys\n\n\ndef has_key(keys, key):\n idx = key % len(keys)\n while keys[idx] is not None:\n if keys[idx] == key:\n return True\n idx = (idx + 1) % len(keys)\n\n return False\n\n\ndef linear_search(numbers, number):\n return number in numbers\n" }, { "path": "python_code/hash_chapter1_reimpl_js.py", "content": "from js_reimpl_common import run_op_chapter1_chapter2\n\n\ndef run_op(keys, op, **kwargs):\n return run_op_chapter1_chapter2(\"chapter1\", None, keys, op, **kwargs)\n\n\ndef create_new(numbers):\n return run_op(None, \"create_new\", array=numbers)\n\n\ndef create_new_broken(numbers):\n return run_op(None, \"create_new_broken\", array=numbers)\n\n\ndef has_key(keys, key):\n return run_op(keys, \"has_key\", key=key)\n\n\ndef linear_search(numbers, key):\n return run_op(None, \"linear_search\", key=key, array=numbers)\n" }, { "path": "python_code/hash_chapter1_reimplementation_test.py", "content": "import random\nimport argparse\n\nimport hash_chapter1_reimpl_js\nimport hash_chapter1_impl\nimport build_autogenerated_chapter1_hash\n\n\ndef get_implementation(is_broken, impl):\n if impl == \"js\":\n module = hash_chapter1_reimpl_js\n elif impl == \"py_ref\":\n module = hash_chapter1_impl\n elif impl == \"py_extracted\":\n module = build_autogenerated_chapter1_hash\n else:\n assert False\n\n return (module.create_new_broken if is_broken else module.create_new, module.has_key)\n\n\ndef run(test_implementation, is_broken, n_inserts):\n MAX_VAL = 5000\n ref_create_new, ref_has_key = get_implementation(is_broken, \"py_ref\")\n test_create_new, test_has_key = get_implementation(is_broken, test_implementation)\n\n numbers = list(set(random.randint(-MAX_VAL, MAX_VAL) for _ in range(n_inserts)))\n\n ref_keys = ref_create_new(numbers)\n test_keys = test_create_new(numbers)\n\n for number in numbers:\n if not is_broken:\n assert ref_has_key(ref_keys, number)\n assert test_has_key(test_keys, number)\n else:\n assert ref_has_key(ref_keys, number) == test_has_key(test_keys, number)\n\n for i in range(n_inserts * 3):\n number = random.randint(-MAX_VAL, MAX_VAL)\n assert ref_has_key(ref_keys, number) == test_has_key(test_keys, number)\n\n\nif __name__ == \"__main__\":\n parser = argparse.ArgumentParser(description='Stress-test chapter1 reimplementation')\n parser.add_argument('--is-broken', action='store_true')\n parser.add_argument('--test-implementation', choices=['py_extracted', 'js'], required=True)\n parser.add_argument('--num-inserts', type=int, default=500)\n args = parser.parse_args()\n\n run(test_implementation=args.test_implementation,\n is_broken=args.is_broken,\n n_inserts=args.num_inserts)\n" }, { "path": "python_code/hash_chapter2_impl.py", "content": "from common import DUMMY, EMPTY\n\n\ndef create_new(from_keys):\n n = len(from_keys)\n hash_codes = [EMPTY for i in range(2 * n)]\n keys = [EMPTY for i in range(2 * n)]\n\n for key in from_keys:\n hash_code = hash(key)\n idx = hash_code % len(keys)\n\n while keys[idx] is not EMPTY:\n if hash_codes[idx] == hash_code and keys[idx] == key:\n break\n idx = (idx + 1) % len(keys)\n\n hash_codes[idx] = hash_code\n keys[idx] = key\n\n return hash_codes, keys\n\n\ndef insert(hash_codes, keys, key):\n hash_code = hash(key)\n idx = hash_code % len(keys)\n\n while hash_codes[idx] is not EMPTY:\n if hash_codes[idx] == hash_code and keys[idx] == key:\n return\n idx = (idx + 1) % len(keys)\n\n hash_codes[idx] = hash_code\n keys[idx] = key\n\n\ndef remove(hash_codes, keys, key):\n hash_code = hash(key)\n idx = hash_code % len(keys)\n\n while hash_codes[idx] is not EMPTY:\n if hash_codes[idx] == hash_code and keys[idx] == key:\n keys[idx] = DUMMY\n return\n idx = (idx + 1) % len(keys)\n\n raise KeyError()\n\n\ndef has_key(hash_codes, keys, key):\n hash_code = hash(key)\n idx = hash_code % len(keys)\n while hash_codes[idx] is not EMPTY:\n if hash_codes[idx] == hash_code and keys[idx] == key:\n return True\n idx = (idx + 1) % len(keys)\n return False\n\n\ndef resize(hash_codes, keys):\n new_hash_codes = [EMPTY for i in range(len(hash_codes) * 2)]\n new_keys = [EMPTY for i in range(len(keys) * 2)]\n for hash_code, key in zip(hash_codes, keys):\n if key is EMPTY or key is DUMMY:\n continue\n idx = hash_code % len(new_keys)\n while new_hash_codes[idx] is not EMPTY:\n idx = (idx + 1) % len(new_keys)\n new_hash_codes[idx] = hash_code\n new_keys[idx] = key\n\n return new_hash_codes, new_keys\n" }, { "path": "python_code/hash_chapter2_impl_test.py", "content": "import unittest\nfrom hash_chapter2_impl import create_new, has_key, insert, remove, resize, DUMMY\nfrom common import generate_random_string\n\n\nclass MyHashTest(unittest.TestCase):\n def test_handcrafted(self):\n expected_len = 6\n hashes, keys = create_new([42, 43, 12])\n\n self.assertEqual(len(hashes), expected_len)\n self.assertEqual(len(keys), expected_len)\n insert(hashes, keys, 42)\n\n self.assertEqual(hashes[42 % expected_len], 42)\n self.assertEqual(keys[42 % expected_len], 42)\n\n self.assertEqual(hashes[43 % expected_len], 43)\n self.assertEqual(keys[43 % expected_len], 43)\n\n self.assertEqual(hashes[42 % expected_len], 42)\n self.assertEqual(keys[42 % expected_len], 42)\n\n self.assertEqual(hashes[12 % expected_len], 42)\n self.assertEqual(keys[12 % expected_len], 42)\n self.assertEqual(hashes[12 % expected_len + 1], 43)\n self.assertEqual(keys[12 % expected_len + 1], 43)\n self.assertEqual(hashes[12 % expected_len + 2], 12)\n self.assertEqual(keys[12 % expected_len + 2], 12)\n\n self.assertTrue(has_key(hashes, keys, 42))\n self.assertTrue(has_key(hashes, keys, 43))\n self.assertTrue(has_key(hashes, keys, 12))\n self.assertFalse(has_key(hashes, keys, 45))\n\n # table: [42, 43, 12, None, None, None]\n insert(hashes, keys, \"\") # hash(\"\") == 0\n self.assertEqual(hashes[3], 0)\n self.assertEqual(keys[3], \"\")\n\n self.assertTrue(has_key(hashes, keys, \"\"))\n self.assertTrue(has_key(hashes, keys, 42))\n\n insert(hashes, keys, \"aba\") # hash(\"aba\") % 6 == 5\n self.assertEqual(hashes[5], hash(\"aba\"))\n self.assertEqual(keys[5], \"aba\")\n\n self.assertTrue(has_key(hashes, keys, 12))\n remove(hashes, keys, 12)\n self.assertFalse(has_key(hashes, keys, 12))\n\n self.assertEqual(hashes[12 % expected_len], 42)\n self.assertEqual(keys[12 % expected_len], 42)\n\n self.assertEqual(keys[12 % expected_len + 2], DUMMY)\n\n with self.assertRaises(KeyError):\n remove(hashes, keys, 12)\n with self.assertRaises(KeyError):\n remove(hashes, keys, 45)\n\n self.assertFalse(has_key(hashes, keys, 12))\n self.assertFalse(has_key(hashes, keys, 45))\n self.assertTrue(has_key(hashes, keys, 42))\n self.assertTrue(has_key(hashes, keys, 43))\n self.assertTrue(has_key(hashes, keys, \"\"))\n self.assertTrue(has_key(hashes, keys, \"aba\"))\n\n insert(hashes, keys, \"abg\")\n self.assertTrue(has_key(hashes, keys, \"abg\"))\n self.assertEqual(hashes[4], hash(\"abg\"))\n self.assertEqual(keys[4], \"abg\")\n hashes, keys = resize(hashes, keys)\n\n self.assertTrue(has_key(hashes, keys, 42))\n self.assertTrue(has_key(hashes, keys, 43))\n self.assertTrue(has_key(hashes, keys, \"\"))\n self.assertTrue(has_key(hashes, keys, \"aba\"))\n self.assertTrue(has_key(hashes, keys, \"abg\"))\n\n self.assertFalse(has_key(hashes, keys, 12))\n self.assertFalse(has_key(hashes, keys, 45))\n\n self.assertEqual(hashes[6], 42)\n self.assertEqual(keys[6], 42)\n self.assertEqual(hashes[7], 43)\n self.assertEqual(keys[7], 43)\n\n self.assertEqual(hashes[0], 0)\n self.assertEqual(keys[0], \"\")\n for h in hashes:\n self.assertTrue(h != 12)\n\n self.assertEqual(hashes[5], hash(\"aba\"))\n self.assertEqual(keys[5], \"aba\")\n\n self.assertEqual(hashes[11], hash(\"abg\"))\n self.assertEqual(keys[11], \"abg\")\n\n def test_all(self):\n n = 10\n initial_keys = [generate_random_string() for _ in range(n)]\n more_keys = [generate_random_string() for _ in range(n // 3)]\n myhashes, mykeys = create_new(initial_keys)\n\n for key in more_keys:\n insert(myhashes, mykeys, key)\n insert(myhashes, mykeys, key)\n\n existing_keys = initial_keys + more_keys\n for key in existing_keys:\n self.assertTrue(has_key(myhashes, mykeys, key))\n\n myhashes, mykeys = resize(myhashes, mykeys)\n\n for key in existing_keys:\n self.assertTrue(has_key(myhashes, mykeys, key))\n\n missing_keys = [generate_random_string() for _ in range(3 * n)]\n for key in set(missing_keys) - set(existing_keys):\n self.assertFalse(has_key(myhashes, mykeys, key))\n with self.assertRaises(KeyError):\n remove(myhashes, mykeys, key)\n\n for key in existing_keys:\n self.assertTrue(has_key(myhashes, mykeys, key))\n remove(myhashes, mykeys, key)\n self.assertFalse(has_key(myhashes, mykeys, key))\n\n for key in more_keys:\n self.assertFalse(has_key(myhashes, mykeys, key))\n insert(myhashes, mykeys, key)\n self.assertTrue(has_key(myhashes, mykeys, key))\n remove(myhashes, mykeys, key)\n self.assertFalse(has_key(myhashes, mykeys, key))\n\n\ndef main():\n unittest.main()\n\n\nif __name__ == \"__main__\":\n main()\n" }, { "path": "python_code/hash_chapter2_reimpl_js.py", "content": "from js_reimpl_common import run_op_chapter1_chapter2\n\n\ndef run_op(hash_codes, keys, op, **kwargs):\n return run_op_chapter1_chapter2(\"chapter2\", hash_codes, keys, op, **kwargs)\n\n\ndef create_new(from_keys):\n return run_op(None, None, \"create_new\", array=from_keys)\n\n\ndef insert(hash_codes, keys, key):\n new_hash_codes, new_keys = run_op(hash_codes, keys, \"insert\", key=key)\n hash_codes[:] = new_hash_codes\n keys[:] = new_keys\n\n\ndef remove(hash_codes, keys, key):\n new_hash_codes, new_keys = run_op(hash_codes, keys, \"remove\", key=key)\n hash_codes[:] = new_hash_codes\n keys[:] = new_keys\n\n\ndef has_key(hash_codes, keys, key):\n return run_op(hash_codes, keys, \"has_key\", key=key)\n\n\ndef resize(hash_codes, keys):\n return run_op(hash_codes, keys, \"resize\")\n" }, { "path": "python_code/hash_chapter2_reimplementation_test.py", "content": "import random\nimport argparse\n\nfrom common import DUMMY, EMPTY, AllKeyValueFactory, IntKeyValueFactory\n\nimport hash_chapter2_reimpl_js\nimport hash_chapter2_impl\nimport build_autogenerated_chapter2\n\nTEST_IMPLEMENTATIONS = {\n 'js_reimpl': hash_chapter2_reimpl_js,\n 'py_extracted': build_autogenerated_chapter2\n}\n\n\ndef verify_same(ref_hash_codes, ref_keys, hash_codes, keys):\n if (ref_hash_codes, ref_keys) != (hash_codes, keys):\n print(\"ORIG SIZES\", len(ref_hash_codes), len(ref_keys))\n print(\"NEW SIZES\", len(hash_codes), len(keys))\n if len(ref_hash_codes) == len(hash_codes) == len(ref_keys) == len(keys):\n size = len(hash_codes)\n print(\"NEW | ORIG\")\n for i in range(size):\n if ref_hash_codes[i] is not EMPTY or hash_codes[i] is not EMPTY:\n print(i, \" \" * 3,\n ref_hash_codes[i], ref_keys[i], \" \" * 3,\n hash_codes[i], keys[i], \" \" * 3)\n\n assert ref_hash_codes == hash_codes and ref_keys == keys\n\n\ndef run(ref_impl, test_impl, n_inserts, key_value_factory, initial_state, extra_checks, verbose):\n SINGLE_REMOVE_CHANCE = 0.3\n\n ref_hash_codes, ref_keys = ref_impl.create_new(initial_state)\n test_hash_codes, test_keys = test_impl.create_new(initial_state)\n\n def vs():\n verify_same(ref_hash_codes, ref_keys, test_hash_codes, test_keys)\n\n vs()\n\n if verbose:\n print(\"Starting test\")\n\n for i in range(n_inserts):\n key_to_insert = key_value_factory.generate_key()\n\n existing_keys = set([k for k in ref_keys if k is not DUMMY and k is not EMPTY])\n fill = sum(1 for k in ref_keys if k is not EMPTY)\n if existing_keys and random.random() < SINGLE_REMOVE_CHANCE:\n key_to_remove = random.choice(list(existing_keys))\n assert ref_impl.has_key(ref_hash_codes, ref_keys, key_to_remove)\n assert test_impl.has_key(test_hash_codes, test_keys, key_to_remove)\n\n ref_impl.remove(ref_hash_codes, ref_keys, key_to_remove)\n test_impl.remove(test_hash_codes, test_keys, key_to_remove)\n existing_keys.remove(key_to_remove)\n\n assert not ref_impl.has_key(ref_hash_codes, ref_keys, key_to_remove)\n assert not test_impl.has_key(test_hash_codes, test_keys, key_to_remove)\n\n is_key_present = ref_impl.has_key(ref_hash_codes, ref_keys, key_to_insert)\n assert (key_to_insert in existing_keys) == is_key_present\n\n if not is_key_present:\n if verbose:\n print(\"Inserting {}\".format(key_to_insert))\n assert not test_impl.has_key(test_hash_codes, test_keys, key_to_insert)\n else:\n if verbose:\n print(\"Re-Inserting {}\".format(key_to_insert))\n\n ref_impl.insert(ref_hash_codes, ref_keys, key_to_insert)\n test_impl.insert(test_hash_codes, test_keys, key_to_insert)\n vs()\n assert test_impl.has_key(test_hash_codes, test_keys, key_to_insert)\n assert ref_impl.has_key(ref_hash_codes, ref_keys, key_to_insert)\n\n if fill / len(ref_keys) > 0.66:\n ref_hash_codes, ref_keys = ref_impl.resize(ref_hash_codes, ref_keys)\n test_hash_codes, test_keys = test_impl.resize(test_hash_codes, test_keys)\n vs()\n\n if extra_checks:\n for k in existing_keys:\n assert test_impl.has_key(test_hash_codes, test_keys, k)\n assert ref_impl.has_key(ref_hash_codes, ref_keys, k)\n\n\nif __name__ == \"__main__\":\n parser = argparse.ArgumentParser(description='Stress-test chapter2 reimplementation')\n parser.add_argument('--test-implementation', choices=TEST_IMPLEMENTATIONS.keys(), required=True)\n parser.add_argument('--num-inserts', type=int, default=500)\n parser.add_argument('--forever', action='store_true')\n parser.add_argument('--kv', choices=[\"numbers\", \"all\"], required=True)\n parser.add_argument('--initial-size', type=int, default=-1)\n parser.add_argument('--extra-getitem-checks', action='store_true', default=False)\n parser.add_argument('--verbose', action='store_true', default=False)\n args = parser.parse_args()\n\n if args.kv == \"numbers\":\n kv_factory = IntKeyValueFactory(args.num_inserts)\n elif args.kv == \"all\":\n kv_factory = AllKeyValueFactory(args.num_inserts)\n\n def test_iteration():\n initial_size = args.initial_size if args.initial_size >= 0 else random.randint(0, 100)\n initial_state = [kv_factory.generate_key() for _ in range(initial_size)]\n run(hash_chapter2_impl,\n TEST_IMPLEMENTATIONS[args.test_implementation],\n n_inserts=args.num_inserts,\n key_value_factory=kv_factory,\n initial_state=initial_state,\n extra_checks=args.extra_getitem_checks,\n verbose=args.verbose)\n\n if args.forever:\n while True:\n test_iteration()\n else:\n test_iteration()\n" }, { "path": "python_code/hash_chapter3_class_impl.py", "content": "from common import DUMMY, EMPTY\nfrom dict_reimpl_common import BaseDictImpl, Slot\n\n\nclass AlmostPythonDictBase(BaseDictImpl):\n START_SIZE = 8\n\n def __init__(self, pairs=None):\n BaseDictImpl.__init__(self)\n self._keys_set = set()\n if pairs:\n for k, v in pairs:\n self[k] = v\n\n def lookdict(self, key):\n hash_code = hash(key)\n\n idx = hash_code % len(self.slots)\n while self.slots[idx].key is not EMPTY:\n if self.slots[idx].hash_code == hash_code and self.slots[idx].key == key:\n return idx\n\n idx = (idx + 1) % len(self.slots)\n\n raise KeyError()\n\n def __getitem__(self, key):\n idx = self.lookdict(key)\n\n return self.slots[idx].value\n\n def __delitem__(self, key):\n idx = self.lookdict(key)\n\n self.used -= 1\n self.slots[idx].key = DUMMY\n self.slots[idx].value = EMPTY\n self._keys_set.remove(key)\n\n def resize(self):\n old_slots = self.slots\n new_size = self.find_nearest_size(2 * self.used)\n self.slots = [Slot() for _ in range(new_size)]\n\n for slot in old_slots:\n if slot.key is not EMPTY and slot.key is not DUMMY:\n idx = slot.hash_code % len(self.slots)\n while self.slots[idx].key is not EMPTY:\n idx = (idx + 1) % len(self.slots)\n\n self.slots[idx] = Slot(slot.hash_code, slot.key, slot.value)\n\n self.fill = self.used\n\n def keys(self):\n return self._keys_set\n\n def __len__(self):\n return len(self.keys())\n\n\nclass AlmostPythonDictImplementationRecycling(AlmostPythonDictBase):\n def __setitem__(self, key, value):\n hash_code = hash(key)\n idx = hash_code % len(self.slots)\n target_idx = None\n while self.slots[idx].key is not EMPTY:\n if self.slots[idx].hash_code == hash_code and self.slots[idx].key == key:\n target_idx = idx\n break\n if target_idx is None and self.slots[idx].key is DUMMY:\n target_idx = idx\n\n idx = (idx + 1) % len(self.slots)\n\n if target_idx is None:\n target_idx = idx\n\n if self.slots[target_idx].key is EMPTY:\n self.used += 1\n self.fill += 1\n elif self.slots[target_idx].key is DUMMY:\n self.used += 1\n\n self.slots[target_idx] = Slot(hash_code, key, value)\n\n if self.fill * 3 >= len(self.slots) * 2:\n self.resize()\n\n self._keys_set.add(key)\n\n\nclass AlmostPythonDictImplementationNoRecycling(AlmostPythonDictBase):\n def __setitem__(self, key, value):\n hash_code = hash(key)\n idx = hash_code % len(self.slots)\n target_idx = None\n while self.slots[idx].key is not EMPTY:\n if self.slots[idx].hash_code == hash_code and\\\n self.slots[idx].key == key:\n target_idx = idx\n break\n idx = (idx + 1) % len(self.slots)\n\n if target_idx is None:\n target_idx = idx\n if self.slots[target_idx].key is EMPTY:\n self.used += 1\n self.fill += 1\n\n self.slots[target_idx] = Slot(hash_code, key, value)\n if self.fill * 3 >= len(self.slots) * 2:\n self.resize()\n\n self._keys_set.add(key)\n\n\nclass AlmostPythonDictImplementationNoRecyclingSimplerVersion(AlmostPythonDictBase):\n def __setitem__(self, key, value):\n hash_code = hash(key)\n idx = hash_code % len(self.slots)\n while self.slots[idx].key is not EMPTY:\n if self.slots[idx].hash_code == hash_code and\\\n self.slots[idx].key == key:\n break\n idx = (idx + 1) % len(self.slots)\n\n if self.slots[idx].key is EMPTY:\n self.used += 1\n self.fill += 1\n\n self.slots[idx] = Slot(hash_code, key, value)\n if self.fill * 3 >= len(self.slots) * 2:\n self.resize()\n\n self._keys_set.add(key)\n" }, { "path": "python_code/hash_chapter3_class_impl_test.py", "content": "import unittest\nfrom common import DUMMY, EMPTY\nfrom hash_chapter3_class_impl import AlmostPythonDictImplementationRecycling, AlmostPythonDictImplementationNoRecycling\n\n\nclass HashDictImplementationTest(unittest.TestCase):\n def test_handcrafted(self):\n d = AlmostPythonDictImplementationRecycling()\n self.assertEqual(len(d.slots), 8)\n\n def assert_contains(i, h, k, v):\n self.assertEqual(d.slots[i].hash_code, h)\n self.assertEqual(d.slots[i].key, k)\n self.assertEqual(d.slots[i].value, v)\n\n d[\"\"] = 1\n d[17] = 2\n d[18] = 3\n self.assertEqual(d[\"\"], 1)\n self.assertEqual(d[17], 2)\n self.assertEqual(d[18], 3)\n\n assert_contains(0, 0, \"\", 1)\n assert_contains(1, 17, 17, 2)\n assert_contains(2, 18, 18, 3)\n\n self.assertEqual(d.fill, 3)\n self.assertEqual(d.used, 3)\n\n with self.assertRaises(KeyError):\n del d[1]\n\n del d[17]\n assert_contains(1, 17, DUMMY, EMPTY)\n\n self.assertEqual(d.fill, 3)\n self.assertEqual(d.used, 2)\n # hash(\"abcd\") % 8 == 0\n\n # py 3.2 hash()\n d[\"abcd\"] = 4\n self.assertEqual(d[\"abcd\"], 4)\n assert_contains(1, -2835746963027601024, \"abcd\", 4)\n self.assertEqual(d.fill, 3)\n self.assertEqual(d.used, 3)\n\n d[\"abcd\"] = 5\n self.assertEqual(d[\"abcd\"], 5)\n assert_contains(1, -2835746963027601024, \"abcd\", 5)\n self.assertEqual(d.fill, 3)\n self.assertEqual(d.used, 3)\n\n del d[\"abcd\"]\n with self.assertRaises(KeyError):\n d[\"abcd\"]\n\n d[15] = 6\n d[14] = 7\n\n assert_contains(7, 15, 15, 6)\n assert_contains(6, 14, 14, 7)\n\n self.assertEqual(len(d.slots), 8)\n self.assertEqual(d.fill, 5)\n self.assertEqual(d.used, 4)\n d[13] = 8\n self.assertEqual(len(d.slots), 16)\n self.assertEqual(d.fill, 5)\n self.assertEqual(d.used, 5)\n\n assert_contains(0, 0, \"\", 1)\n assert_contains(2, 18, 18, 3)\n assert_contains(13, 13, 13, 8)\n assert_contains(14, 14, 14, 7)\n assert_contains(15, 15, 15, 6)\n\n def test_handcrafted_simple_setitem(self):\n d = AlmostPythonDictImplementationNoRecycling()\n self.assertEqual(len(d.slots), 8)\n\n def assert_contains(i, h, k, v):\n self.assertEqual(d.slots[i].hash_code, h)\n self.assertEqual(d.slots[i].key, k)\n self.assertEqual(d.slots[i].value, v)\n\n d[\"\"] = 1\n d[17] = 2\n d[18] = 3\n self.assertEqual(d[\"\"], 1)\n self.assertEqual(d[17], 2)\n self.assertEqual(d[18], 3)\n\n assert_contains(0, 0, \"\", 1)\n assert_contains(1, 17, 17, 2)\n assert_contains(2, 18, 18, 3)\n\n self.assertEqual(d.fill, 3)\n self.assertEqual(d.used, 3)\n\n with self.assertRaises(KeyError):\n del d[1]\n\n del d[17]\n assert_contains(1, 17, DUMMY, EMPTY)\n\n self.assertEqual(d.fill, 3)\n self.assertEqual(d.used, 2)\n # hash(\"abcd\") % 8 == 0\n\n # py 3.2 hash()\n d[\"abcd\"] = 4\n self.assertEqual(d[\"abcd\"], 4)\n assert_contains(3, -2835746963027601024, \"abcd\", 4)\n self.assertEqual(d.fill, 4)\n self.assertEqual(d.used, 3)\n\n d[\"abcd\"] = 5\n self.assertEqual(d[\"abcd\"], 5)\n assert_contains(3, -2835746963027601024, \"abcd\", 5)\n self.assertEqual(d.fill, 4)\n self.assertEqual(d.used, 3)\n\n del d[\"abcd\"]\n with self.assertRaises(KeyError):\n d[\"abcd\"]\n\n self.assertEqual(len(d.slots), 8)\n self.assertEqual(d.fill, 4)\n self.assertEqual(d.used, 2)\n\n d[15] = 6\n self.assertEqual(len(d.slots), 8)\n self.assertEqual(d.fill, 5)\n self.assertEqual(d.used, 3)\n assert_contains(7, 15, 15, 6)\n\n d[13] = 8\n self.assertEqual(len(d.slots), 16)\n self.assertEqual(d.fill, 4)\n self.assertEqual(d.used, 4)\n\n assert_contains(0, 0, \"\", 1)\n assert_contains(2, 18, 18, 3)\n assert_contains(13, 13, 13, 8)\n assert_contains(15, 15, 15, 6)\n\n\ndef main():\n unittest.main()\n\n\nif __name__ == \"__main__\":\n main()\n" }, { "path": "python_code/interface_test.py", "content": "import unittest\nfrom dict32_reimplementation import PyDictReimplementation\nfrom hash_chapter3_class_impl import AlmostPythonDictImplementationRecycling, AlmostPythonDictImplementationNoRecycling\nfrom js_reimplementation_interface import Dict32JsImpl, AlmostPythonDictRecyclingJsImpl, AlmostPythonDictNoRecyclingJsImpl\n\n\nclass Interface(unittest.TestCase):\n def test_all(self):\n self.do_simple_test_single_class(PyDictReimplementation)\n self.do_simple_test_single_class(AlmostPythonDictImplementationRecycling)\n self.do_simple_test_single_class(AlmostPythonDictImplementationNoRecycling)\n\n self.do_simple_test_single_class(Dict32JsImpl)\n self.do_simple_test_single_class(AlmostPythonDictRecyclingJsImpl)\n self.do_simple_test_single_class(AlmostPythonDictNoRecyclingJsImpl)\n\n def do_simple_test_single_class(self, klass):\n d = klass()\n\n for i in range(100):\n d[i] = i\n self.assertEqual(d[i], i)\n\n for i in range(50):\n del d[i]\n with self.assertRaises(KeyError):\n d[i]\n\n for i in range(200):\n d[i] = i + 1\n self.assertEqual(d[i], i + 1)\n\n\ndef main():\n unittest.main()\n\n\nif __name__ == \"__main__\":\n main()\n" }, { "path": "python_code/js_reimpl_common.py", "content": "import socket\nimport json\n\nfrom common import DUMMY, EMPTY\n\nnone_info = {\n \"type\": \"None\",\n \"hash\": str(hash(None))\n}\n\n\ndef dump_simple_py_obj(obj):\n if obj is DUMMY:\n return {\n \"type\": \"DUMMY\"\n }\n elif obj is EMPTY:\n return None\n elif obj is None:\n return none_info\n elif isinstance(obj, int):\n return {\n 'type': 'int',\n 'value': str(obj)\n }\n return obj\n\n\ndef dump_pairs(pairs):\n res = []\n for k, v in pairs:\n res.append([dump_simple_py_obj(k), dump_simple_py_obj(v)])\n\n return res\n\n\ndef dump_array(array):\n return list(map(dump_simple_py_obj, array))\n\n\ndef parse_array(array):\n return list(map(parse_simple_py_obj, array))\n\n\ndef parse_simple_py_obj(obj):\n if isinstance(obj, dict):\n assert obj[\"type\"] in [\"DUMMY\", \"None\", \"int\"]\n if obj[\"type\"] == \"DUMMY\":\n return DUMMY\n if obj[\"type\"] == \"None\":\n return None\n return int(obj[\"value\"])\n elif obj is None:\n return EMPTY\n return obj\n\n\nsock = None\nsockfile = None\n\n\ndef _init_sock_stuff():\n global sock\n global sockfile\n\n # TODO: unhardcode?\n SOCK_FILENAME = 'pynode.sock'\n\n if sock is None:\n sock = socket.socket(socket.AF_UNIX, socket.SOCK_STREAM)\n sock.connect(SOCK_FILENAME)\n sockfile = sock.makefile('r')\n\n return sock, sockfile\n\n\ndef run_op_chapter1_chapter2(chapter, hash_codes, keys, op, **kwargs):\n _init_sock_stuff()\n\n for name in kwargs:\n if name != 'array':\n kwargs[name] = dump_simple_py_obj(kwargs[name])\n else:\n kwargs[name] = dump_array(kwargs[name])\n\n data = {\n \"dict\": chapter,\n \"op\": op,\n \"args\": kwargs,\n \"hashCodes\": dump_array(hash_codes) if hash_codes is not None else None,\n \"keys\": dump_array(keys) if keys is not None else None,\n }\n\n sock.send(bytes(json.dumps(data) + \"\\n\", 'UTF-8'))\n response = json.loads(sockfile.readline())\n\n if \"exception\" in response and response[\"exception\"]:\n raise KeyError()\n\n if 'result' in response and response['result'] is not None:\n # TODO: this is pretty hacky\n return response[\"result\"]\n elif \"hashCodes\" in response:\n return parse_array(response[\"hashCodes\"]), parse_array(response[\"keys\"])\n else:\n return parse_array(response[\"keys\"])\n" }, { "path": "python_code/js_reimplementation_interface.py", "content": "import socket\nimport json\n\nfrom common import DUMMY, EMPTY\nfrom js_reimpl_common import dump_simple_py_obj, parse_simple_py_obj, dump_pairs\nfrom dict_reimpl_common import Slot\n\n\nclass JsImplBase(object):\n # TODO: unhardcode?\n SOCK_FILENAME = 'pynode.sock'\n\n def __init__(self, pairs=None):\n pairs = pairs or []\n\n self.sock = socket.socket(socket.AF_UNIX, socket.SOCK_STREAM)\n self.sock.connect(self.SOCK_FILENAME)\n self.sockfile = self.sock.makefile('r')\n\n self.slots = None\n self.fill = None\n self.used = None\n\n self.run_op(\"__init__\", pairs=pairs)\n\n def __del__(self):\n self.sock.close()\n\n def dump_slots(self):\n def dump_slot(slot):\n key = dump_simple_py_obj(slot.key)\n value = dump_simple_py_obj(slot.value)\n\n hash_code = slot.hash_code\n if hash_code is EMPTY:\n hash_code = None\n\n return {\n \"hashCode\": str(hash_code) if hash_code is not None else None,\n \"key\": key,\n \"value\": value,\n }\n\n if self.slots is None:\n return None\n\n return list(map(dump_slot, self.slots))\n\n def restore_slots(self, slots):\n def restore_slot(slot):\n key = parse_simple_py_obj(slot[\"key\"])\n value = parse_simple_py_obj(slot[\"value\"])\n assert value is not DUMMY\n\n hash_code = int(slot[\"hashCode\"]) if slot[\"hashCode\"] is not None else None\n if hash_code is None:\n hash_code = EMPTY\n\n return Slot(hash_code, key, value)\n\n self.slots = list(map(restore_slot, slots))\n\n def run_op(self, op, **kwargs):\n for name in kwargs:\n if name != 'pairs':\n kwargs[name] = dump_simple_py_obj(kwargs[name])\n else:\n kwargs[name] = dump_pairs(kwargs[name])\n\n data = {\n \"dict\": self.dict_type,\n \"op\": op,\n \"args\": kwargs,\n \"self\": {\n \"slots\": self.dump_slots(),\n \"used\": self.used,\n \"fill\": self.fill\n }\n }\n\n # pprint((\"<< sending\", data, op, kwargs))\n self.sock.send(bytes(json.dumps(data) + \"\\n\", 'UTF-8'))\n response = json.loads(self.sockfile.readline())\n # pprint((\">> receiving\", response))\n\n self.restore_slots(response[\"self\"][\"slots\"])\n self.fill = response[\"self\"][\"fill\"]\n self.used = response[\"self\"][\"used\"]\n if response[\"exception\"]:\n raise KeyError(\"whatever\")\n\n return parse_simple_py_obj(response[\"result\"])\n\n\nclass Dict32JsImpl(JsImplBase):\n dict_type = \"dict32\"\n\n def __setitem__(self, key, value):\n return self.run_op(\"__setitem__\", key=key, value=value)\n\n def __delitem__(self, key):\n return self.run_op(\"__delitem__\", key=key)\n\n def __getitem__(self, key):\n return self.run_op(\"__getitem__\", key=key)\n\n\nclass AlmostPythonDictBaseJsImpl(JsImplBase):\n dict_type = \"almost_python_dict\"\n\n def __delitem__(self, key):\n return self.run_op(\"__delitem__\", key=key)\n\n def __getitem__(self, key):\n return self.run_op(\"__getitem__\", key=key)\n\n\nclass AlmostPythonDictRecyclingJsImpl(AlmostPythonDictBaseJsImpl):\n def __setitem__(self, key, value):\n return self.run_op(\"__setitem__recycling\", key=key, value=value)\n\n\nclass AlmostPythonDictNoRecyclingJsImpl(AlmostPythonDictBaseJsImpl):\n def __setitem__(self, key, value):\n return self.run_op(\"__setitem__no_recycling\", key=key, value=value)\n" }, { "path": "scripts/extractPythonCode.js", "content": "import 'ignore-styles';\n\nimport {\n DICT32_INIT,\n DICT32_SETITEM,\n DICT32_RESIZE_CODE,\n _DICT32_GETITEM_ONLY,\n _DICT32_DELITEM_ONLY,\n DICT32_LOOKDICT,\n STATICMETHOD_SIGNED_TO_UNSIGNED,\n PROBING_PYTHON_CODE,\n} from '../src/chapter4_real_python_dict';\n\nimport {\n HASH_CLASS_INIT_CODE,\n HASH_CLASS_SETITEM_RECYCLING_CODE,\n HASH_CLASS_SETITEM_SIMPLIFIED_CODE,\n _HASH_CLASS_GETITEM_ONLY,\n _HASH_CLASS_DELITEM_ONLY,\n HASH_CLASS_LOOKDICT,\n HASH_CLASS_RESIZE_CODE,\n FIND_NEAREST_SIZE_CODE_STRING,\n SLOT_CLASS_CODE_STRING,\n} from '../src/chapter3_hash_class';\n\nimport {\n HASH_CREATE_NEW_CODE,\n HASH_SEARCH_CODE,\n HASH_REMOVE_CODE,\n HASH_RESIZE_CODE,\n HASH_INSERT_CODE,\n} from '../src/chapter2_hash_table_functions';\n\nimport {\n SIMPLIFIED_INSERT_ALL_BROKEN_CODE,\n SIMPLIFIED_INSERT_ALL_CODE,\n SIMPLIFIED_SEARCH_CODE,\n SIMPLE_LIST_SEARCH,\n} from '../src/chapter1_simplified_hash';\n\nimport fs from 'fs';\nimport * as path from 'path';\n\nfunction extractCodeLines(codeWithBpAndLevels) {\n return codeWithBpAndLevels.map(([line, bp, level]) => line);\n}\n\nfunction outputCode(filename, headers, importedCode, indent4 = true) {\n let allLines = [];\n for (let part of importedCode) {\n let lines;\n if (typeof part !== 'string') {\n lines = extractCodeLines(part);\n } else {\n lines = part.split('\\n');\n }\n\n allLines.push(...lines);\n if (lines[lines.length - 1] !== '') {\n allLines.push('');\n }\n }\n const joinedLines = allLines.map(line => (line.length > 0 && indent4 ? ' ' + line : line)).join('\\n');\n fs.writeFileSync(filename, headers.join('\\n') + '\\n' + joinedLines);\n}\n\nconst commonImports = `import sys\nimport os\nsys.path.append(os.path.join(os.path.dirname(os.path.abspath(__file__)), '..', 'python_code'))\nfrom common import DUMMY, EMPTY\n\n`;\n\nconst dict32def = `\nclass Dict32Extracted(object):`;\n\nconst DIR = 'build';\n\noutputCode(\n path.join(DIR, 'dict32js_extracted.py'),\n [commonImports, SLOT_CLASS_CODE_STRING, dict32def],\n [\n DICT32_INIT,\n FIND_NEAREST_SIZE_CODE_STRING,\n STATICMETHOD_SIGNED_TO_UNSIGNED,\n DICT32_SETITEM,\n DICT32_RESIZE_CODE,\n _DICT32_GETITEM_ONLY,\n _DICT32_DELITEM_ONLY,\n DICT32_LOOKDICT,\n ]\n);\n\nconst hashClassRecyclingDef = `\nclass HashClassRecyclingExtracted(object):`;\n\noutputCode(\n path.join(DIR, 'hash_class_recycling_extracted.py'),\n [commonImports, SLOT_CLASS_CODE_STRING, hashClassRecyclingDef],\n [\n HASH_CLASS_INIT_CODE,\n FIND_NEAREST_SIZE_CODE_STRING,\n HASH_CLASS_SETITEM_RECYCLING_CODE,\n HASH_CLASS_RESIZE_CODE,\n _HASH_CLASS_GETITEM_ONLY,\n _HASH_CLASS_DELITEM_ONLY,\n HASH_CLASS_LOOKDICT,\n ]\n);\n\nconst hashClassNoRecyclingDef = `\nclass HashClassNoRecyclingExtracted(object):`;\n\noutputCode(\n path.join(DIR, 'hash_class_no_recycling_extracted.py'),\n [commonImports, SLOT_CLASS_CODE_STRING, hashClassNoRecyclingDef],\n [\n HASH_CLASS_INIT_CODE,\n FIND_NEAREST_SIZE_CODE_STRING,\n HASH_CLASS_SETITEM_SIMPLIFIED_CODE,\n HASH_CLASS_RESIZE_CODE,\n _HASH_CLASS_GETITEM_ONLY,\n _HASH_CLASS_DELITEM_ONLY,\n HASH_CLASS_LOOKDICT,\n ]\n);\n\noutputCode(\n path.join(DIR, 'hash_chapter2_extracted.py'),\n [commonImports],\n [HASH_CREATE_NEW_CODE, HASH_SEARCH_CODE, HASH_REMOVE_CODE, HASH_RESIZE_CODE, HASH_INSERT_CODE],\n false\n);\n\noutputCode(\n path.join(DIR, 'hash_chapter1_extracted.py'),\n [commonImports],\n [SIMPLIFIED_INSERT_ALL_CODE, SIMPLIFIED_INSERT_ALL_BROKEN_CODE, SIMPLIFIED_SEARCH_CODE, SIMPLE_LIST_SEARCH],\n false\n);\n\noutputCode(path.join(DIR, 'chapter4_probing_python_code.py'), [commonImports], [PROBING_PYTHON_CODE], false);\n" }, { "path": "scripts/pyReimplWrapper.js", "content": "const net = require('net');\nconst split = require('split');\nimport 'ignore-styles';\n\nimport {BigNumber} from 'bignumber.js';\nimport {DUMMY, EMPTY, None} from '../src/hash_impl_common';\nimport {Dict32} from '../src/chapter4_real_python_dict';\nimport {GenerateProbingLinks} from '../src/probing_visualization.js';\nimport {AlmostPythonDict} from '../src/chapter3_hash_class';\nimport {Ops as Chapter2Ops} from '../src/chapter2_hash_table_functions';\nimport {Ops as Chapter1Ops} from '../src/chapter1_simplified_hash';\nimport {Slot} from '../src/chapter3_and_4_common';\nimport {List as ImmutableList} from 'immutable';\n\nfunction parseSimplePyObj(obj) {\n if (obj === null || typeof obj === 'string') {\n return obj;\n } else if (typeof obj === 'object' && obj.type === 'None') {\n let res = None;\n // TODO FIXME: this does not support multiple clients\n res._hashCode = obj.hash;\n return res;\n } else if (typeof obj === 'object' && obj.type === 'DUMMY') {\n return DUMMY;\n } else if (typeof obj === 'object' && obj.type === 'EMPTY') {\n return EMPTY;\n } else if (typeof obj === 'object' && obj.type === 'int') {\n return BigNumber(obj.value);\n } else {\n throw new Error(`Unknown obj ${JSON.stringify(obj)}`);\n }\n}\n\nfunction parseArray(array) {\n return array.map(parseSimplePyObj);\n}\n\nfunction dumpArray(array) {\n return array.map(dumpSimplePyObj);\n}\n\nfunction parsePairs(pairs) {\n return pairs.map(([k, v]) => [parseSimplePyObj(k), parseSimplePyObj(v)]);\n}\n\nfunction dumpSimplePyObj(obj) {\n if (obj === DUMMY) {\n return {\n type: 'DUMMY',\n };\n } else if (obj === None) {\n return {\n type: 'None',\n };\n } else if (BigNumber.isBigNumber(obj)) {\n return {\n type: 'int',\n value: obj.toString(),\n };\n } else {\n return obj;\n }\n}\n\nfunction restorePyDictState(state) {\n let {pySelf} = Dict32.__init__();\n if (state.slots != null) {\n pySelf = pySelf.set(\n 'slots',\n new ImmutableList(\n state.slots.map(slot => {\n let key = parseSimplePyObj(slot.key);\n let value = parseSimplePyObj(slot.value);\n\n return Slot({\n pyHashCode: slot.hashCode ? new BigNumber(slot.hashCode) : null,\n key: key,\n value: value,\n });\n })\n )\n );\n } else {\n pySelf = pySelf.set('slots', null);\n }\n pySelf = pySelf.set('used', state.used);\n pySelf = pySelf.set('fill', state.fill);\n\n return pySelf;\n}\n\nfunction dumpPyDictState(pySelf) {\n let data = {};\n\n data.slots = pySelf.get('slots').map(slot => {\n return {\n hashCode: slot.pyHashCode != null ? slot.pyHashCode.toString() : null,\n key: dumpSimplePyObj(slot.key),\n value: dumpSimplePyObj(slot.value),\n };\n });\n data.used = pySelf.get('used');\n data.fill = pySelf.get('fill');\n\n return data;\n}\n\nfunction dict32RunOp(pySelf, op, key, value, pairs) {\n switch (op) {\n case '__init__':\n pySelf = Dict32.__init__(pairs).pySelf;\n return {pySelf};\n case '__getitem__': {\n const {result, isException} = Dict32.__getitem__(pySelf, key);\n return {pySelf, result, isException};\n }\n case '__setitem__': {\n ({pySelf} = Dict32.__setitem__(pySelf, key, value));\n return {pySelf};\n }\n case '__delitem__': {\n let isException;\n ({pySelf, isException} = Dict32.__delitem__(pySelf, key));\n return {pySelf, isException};\n }\n default:\n throw new Error('Unknown op: ' + op);\n }\n}\n\nfunction almostPyDictRunOp(pySelf, op, key, value, pairs) {\n switch (op) {\n case '__init__':\n pySelf = AlmostPythonDict.__init__(pairs).pySelf;\n return {pySelf};\n case '__getitem__': {\n const {result, isException} = AlmostPythonDict.__getitem__(pySelf, key);\n return {pySelf, result, isException};\n }\n case '__setitem__recycling': {\n ({pySelf} = AlmostPythonDict.__setitem__recycling(pySelf, key, value));\n return {pySelf};\n }\n case '__setitem__no_recycling': {\n ({pySelf} = AlmostPythonDict.__setitem__no_recycling(pySelf, key, value));\n return {pySelf};\n }\n case '__delitem__': {\n let isException;\n ({pySelf, isException} = AlmostPythonDict.__delitem__(pySelf, key));\n return {pySelf, isException};\n }\n default:\n throw new Error('Unknown op: ' + op);\n }\n}\n\nfunction chapter1run(keys, op, key, numbers) {\n switch (op) {\n case 'create_new':\n ({keys} = Chapter1Ops.createNew(numbers));\n return {keys};\n case 'create_new_broken':\n ({keys} = Chapter1Ops.createNewBroken(numbers));\n return {keys};\n case 'has_key': {\n let result;\n ({keys, result} = Chapter1Ops.hasKey(keys, key));\n return {keys, result};\n }\n case 'linear_search': {\n let {result} = Chapter1Ops.linearSearch(numbers, key);\n return {result};\n }\n default:\n throw new Error('Unknown op: ' + op);\n }\n}\n\nfunction chapter2run(hashCodes, keys, op, key, array) {\n switch (op) {\n case 'create_new':\n ({hashCodes, keys} = Chapter2Ops.createNew(array));\n return {hashCodes, keys};\n case 'insert':\n ({hashCodes, keys} = Chapter2Ops.insert(hashCodes, keys, key));\n return {hashCodes, keys};\n case 'remove': {\n let isException;\n ({hashCodes, keys, isException} = Chapter2Ops.remove(hashCodes, keys, key));\n return {hashCodes, keys, isException};\n }\n case 'has_key': {\n let result;\n ({hashCodes, keys, result} = Chapter2Ops.hasKey(hashCodes, keys, key));\n return {hashCodes, keys, result};\n }\n case 'resize':\n ({hashCodes, keys} = Chapter2Ops.resize(hashCodes, keys));\n return {hashCodes, keys};\n default:\n throw new Error('Unknown op: ' + op);\n }\n}\n\nconst server = net.createServer(c => {\n console.log('Client connected');\n\n c.on('end', () => {\n console.log('Client disconnected');\n });\n\n c.pipe(split()).on('data', line => {\n console.log('Received line of length ' + line.length);\n if (!line) return;\n\n const data = JSON.parse(line);\n const dictType = data.dict;\n const op = data.op;\n let {key, value, pairs, array} = data.args;\n if (key !== undefined) {\n key = parseSimplePyObj(key);\n }\n if (value !== undefined) {\n value = parseSimplePyObj(value);\n }\n if (pairs !== undefined) {\n pairs = parsePairs(pairs);\n }\n if (array !== undefined) {\n array = parseArray(array);\n }\n\n console.log(op, data.args);\n\n let isException, result;\n let response;\n\n if (dictType === 'dict32' || dictType === 'almost_python_dict') {\n let pySelf = restorePyDictState(data.self);\n if (dictType === 'dict32') {\n ({pySelf, isException, result} = dict32RunOp(pySelf, op, key, value, pairs));\n } else if (dictType === 'almost_python_dict') {\n ({pySelf, isException, result} = almostPyDictRunOp(pySelf, op, key, value, pairs));\n } else {\n throw new Error('Unknown dict type');\n }\n\n response = {\n exception: isException || false,\n result: result !== undefined ? dumpSimplePyObj(result) : null,\n self: dumpPyDictState(pySelf),\n };\n } else if (dictType === 'chapter2') {\n let hashCodes = data.hashCodes != null ? new ImmutableList(parseArray(data.hashCodes)) : undefined;\n let keys = data.keys != null ? new ImmutableList(parseArray(data.keys)) : undefined;\n ({hashCodes, keys, isException, result} = chapter2run(hashCodes, keys, op, key, array));\n response = {\n exception: isException || false,\n result: result !== undefined ? result : null,\n hashCodes: dumpArray(hashCodes),\n keys: dumpArray(keys),\n };\n } else if (dictType === 'chapter1') {\n let keys = data.keys != null ? new ImmutableList(parseArray(data.keys)) : undefined;\n ({keys, result} = chapter1run(keys, op, key, array));\n response = {\n result: result !== undefined ? result : null,\n keys: keys !== undefined ? dumpArray(keys) : null,\n };\n } else if (dictType === 'pythonProbing') {\n let g = new GenerateProbingLinks();\n const result = g.run(data.args.slotsCount, key, 'python');\n response = {\n result,\n };\n } else {\n throw new Error('Unknown dict type');\n }\n\n c.write(JSON.stringify(response) + '\\n');\n });\n});\n\nserver.on('error', err => {\n throw err;\n});\n\nserver.on('listening', () => {\n console.log(`Listening`);\n});\n\nserver.listen('pynode.sock', () => {\n console.log('Starting listening...');\n});\n" }, { "path": "scripts/ssr.js", "content": "require('dotenv').config();\n\nprocess.env.NODE_ENV = 'ssr';\n\nconsole.log = () => {}; // Do not log to stdout\nglobal.performance = {now: () => 0};\nimport 'ignore-styles';\n\nimport * as React from 'react';\nimport ReactDOMServer from 'react-dom/server';\nimport {CHAPTER_ID_TO_COMPONENT} from '../src/index';\nimport {App} from '../src/app';\nimport fs from 'fs';\n\nconst filename = process.argv[2];\nconst chapterIds = JSON.parse(process.argv[3]);\nconst chapters = chapterIds.map(id => CHAPTER_ID_TO_COMPONENT[id]);\nlet selectedChapterId;\nif (chapterIds.length === 1) {\n selectedChapterId = chapterIds[0];\n}\n\nfs.readFile(filename, 'utf8', function(err, file) {\n if (err) {\n throw new Error(`Cannot read source html: ${err}`);\n }\n const renderedComponent = ReactDOMServer.renderToString(\n \n );\n let fullHtml = file.replace(/
<\\/div>/, `
${renderedComponent}
`);\n const gaId = process.env.GA_ID;\n console.warn('Google analytics ID is', gaId);\n if (gaId) {\n let GA_SCRIPT = `\n \n `;\n GA_SCRIPT = GA_SCRIPT.replace(/__GA_CODE_HERE__/g, gaId);\n fullHtml = fullHtml.replace('', `${GA_SCRIPT}`);\n }\n process.stdout.write(fullHtml);\n});\n" }, { "path": "src/app.js", "content": "import _ from 'lodash';\nimport Bootstrap from 'bootstrap/dist/css/bootstrap.min.css';\nimport stylesCss from './styles.css';\n\nimport * as React from 'react';\nimport ReactDOM from 'react-dom';\n\nimport {MyErrorBoundary, initUxSettings, getUxSettings, BootstrapAlert, doubleRAF} from './util';\nimport {win, globalSettings} from './store';\n\nimport {faDesktop} from '@fortawesome/free-solid-svg-icons/faDesktop';\nimport {faSpinner} from '@fortawesome/free-solid-svg-icons/faSpinner';\nimport {faSyncAlt} from '@fortawesome/free-solid-svg-icons/faSyncAlt';\nimport {faEnvelope} from '@fortawesome/free-solid-svg-icons/faEnvelope';\nimport {faChevronRight} from '@fortawesome/free-solid-svg-icons/faChevronRight';\nimport {faFirefox} from '@fortawesome/free-brands-svg-icons/faFirefox';\nimport {faGithub} from '@fortawesome/free-brands-svg-icons/faGithub';\nimport {faTwitter} from '@fortawesome/free-brands-svg-icons/faTwitter';\nimport {faMailchimp} from '@fortawesome/free-brands-svg-icons/faMailchimp';\n\nimport {library, config as fontAwesomeConfig} from '@fortawesome/fontawesome-svg-core';\nfontAwesomeConfig.autoAddCss = false;\n\nlibrary.add(faDesktop);\nlibrary.add(faFirefox);\nlibrary.add(faSpinner);\nlibrary.add(faSyncAlt);\nlibrary.add(faEnvelope);\nlibrary.add(faChevronRight);\nlibrary.add(faGithub);\nlibrary.add(faMailchimp);\nlibrary.add(faTwitter);\n\nimport {FontAwesomeIcon} from '@fortawesome/react-fontawesome';\n\nimport '@fortawesome/fontawesome-svg-core/styles.css';\n\nfunction getWindowDimensions() {\n const width = document.documentElement.clientWidth;\n const height = document.documentElement.clientHeight;\n return {width, height};\n}\n\nfunction logViewportStats() {\n console.log(`DIMENSIONS: window inner: ${window.innerWidth}x${window.innerHeight}`);\n console.log(\n `DIMENSIONS: document.documentElement: ${document.documentElement.clientWidth}x${\n document.documentElement.clientHeight\n }`\n );\n const vv = window.visualViewport;\n console.log(`DIMENSIONS: visualViewport: ${vv != null ? vv.width + 'x' + vv.height : vv}`);\n\n const {width, height} = getWindowDimensions();\n console.log(`DIMENSIONS: used: ${width}x${height}`);\n // TODO FIXME: this is for debugging only\n /*const url = `/viewports?wi=${window.innerWidth}x${window.innerHeight}&de=${document.documentElement.clientWidth}x${document.documentElement.clientHeight}&vv=${vv.width}x${vv.height}`;\n const Http = new XMLHttpRequest();\n Http.open(\"GET\", url);\n Http.send();*/\n}\n\nconst GITHUB_REPO_URL = 'https://github.com/eleweek/inside_python_dict';\nconst MAILCHIMP_URL = 'http://eepurl.com/gbzhvn';\nconst TWITTER_LINK = 'https://twitter.com/SashaPutilin';\nconst EMAIL = 'avp-13@yandex.ru';\n\nfunction GithubRibbon() {\n return (\n \n \n \n );\n}\n\nfunction GithubCorner() {\n // FROM: http://tholman.com/github-corners/\n return (\n `,\n }}\n />\n );\n}\n\nfunction GithubForkMe({windowWidth}) {\n /*if (windowWidth != null && windowWidth > 1150) {\n return ;\n } else {*/\n return ;\n /*}*/\n}\n\nconst CONTENTS_DATA = [\n [1, 'chapter1.html', 'Searching efficiently in a list'],\n [2, 'chapter2.html', 'Why are hash tables called hash tables?'],\n [3, 'chapter3.html', 'Putting it all together to make an \"almost\"-python-dict'],\n [4, 'chapter4.html', 'How python dict *really* works internally'],\n];\n\nfunction chapterIdDotHtml(chapterId) {\n if (chapterId && !chapterId.endsWith('.html')) {\n return chapterId + '.html';\n } else {\n return null;\n }\n}\n\nfunction NextPrev({selectedChapterId}) {\n const selectedChapter = chapterIdDotHtml(selectedChapterId);\n if (selectedChapter == null) {\n return null;\n }\n\n let prevHref, prevTitle;\n let nextHref, nextTitle;\n\n for (let i = 0; i < CONTENTS_DATA.length; ++i) {\n if (CONTENTS_DATA[i][1] === selectedChapter) {\n if (i > 0) {\n prevHref = CONTENTS_DATA[i - 1][1];\n prevTitle = CONTENTS_DATA[i - 1][2];\n }\n if (i < CONTENTS_DATA.length - 1) {\n nextHref = CONTENTS_DATA[i + 1][1];\n nextTitle = CONTENTS_DATA[i + 1][2];\n }\n break;\n }\n }\n\n if (nextHref) {\n return (\n \n );\n } else {\n return null;\n }\n}\n\nclass Contents extends React.PureComponent {\n static EXTRA_ERROR_BOUNDARY = true;\n\n render() {\n const {selectedChapterId} = this.props;\n const selectedChapter = chapterIdDotHtml(selectedChapterId);\n const CIRCLE_SIZE = 30;\n return (\n
\n
\n {CONTENTS_DATA.map(([i, href, title]) => {\n const contentRow = (\n \n \n {i}\n
\n
\n
{title}
\n
\n \n );\n return (\n \n {selectedChapter === href ? (\n contentRow\n ) : (\n \n {contentRow}\n \n )}\n
\n );\n })}\n
\n \n );\n }\n}\n\nclass LoadingAlert extends React.PureComponent {\n constructor() {\n super();\n\n this.state = {\n loaded: false,\n };\n }\n\n render() {\n return (\n \n JavaScript code is loading...\n \n );\n }\n\n componentDidMount() {\n this.setState({loaded: true});\n }\n}\n\nclass Alerts extends React.Component {\n constructor() {\n super();\n\n this.state = {\n mounted: false,\n };\n }\n\n render() {\n const alerts = [];\n const isRunningInBrowser = typeof window !== 'undefined';\n alerts.push();\n\n if (this.state.mounted) {\n const {browser, windowWidth, windowHeight} = this.props;\n if (browser) {\n if (browser.platform.type === 'mobile') {\n alerts.push(\n \n Mobile device detected. For the best\n experience use a desktop browser\n \n );\n if (windowWidth < windowHeight) {\n alerts.push(\n \n Rotating your device is recommended{' '}\n - animations are better with a wider viewport\n \n );\n }\n } else if (browser.browser.name === 'Firefox' && browser.os.name !== 'Linux') {\n alerts.push(\n \n Firefox detected. Heavy\n animations may lag sometimes. If this happens, Chrome or Safari is recommended.\n \n );\n }\n }\n }\n\n return {alerts};\n }\n\n componentDidMount() {\n this.setState({mounted: true});\n }\n}\n\nfunction Footer() {\n return (\n \n );\n}\n\n// mainly to prevent addressbar stuff on mobile changing things excessively\nconst SIGNIFICANT_HEIGHT_CHANGE = 100;\nexport class App extends React.Component {\n constructor() {\n super();\n this.state = {\n mounted: false,\n windowWidth: null,\n windowHeight: null,\n };\n }\n\n windowSizeChangeHandle = () => {\n logViewportStats();\n const dimensions = getWindowDimensions();\n const windowWidth = dimensions.width;\n const windowHeight = dimensions.height;\n if (this.state.windowWidth !== windowWidth || this.state.windowHeight !== windowHeight) {\n console.log('Processing window size change', windowWidth, windowHeight);\n if (\n this.state.windowWidth != windowWidth ||\n this.state.windowHeight > windowHeight ||\n windowHeight - this.state.windowHeight > SIGNIFICANT_HEIGHT_CHANGE\n ) {\n console.log('App size changed from', this.state);\n this.setState({\n windowWidth,\n windowHeight,\n });\n if (win.width !== windowWidth || win.height !== windowHeight) {\n win.setWH(windowWidth, windowHeight);\n }\n }\n fixStickyResize(windowWidth, windowHeight);\n }\n };\n\n componentDidMount() {\n const MEANINGFUL_Y_DIFF = 50; // components that depend on scroll should allow some leeway\n let lastScrollY = null;\n const onScroll = _.throttle(() => {\n if (!lastScrollY || Math.abs(lastScrollY - window.scrollY) > MEANINGFUL_Y_DIFF) {\n console.log('onScroll triggered', window.scrollY);\n win.setScrollY(window.scrollY);\n lastScrollY = window.scrollY;\n }\n }, 100);\n window.addEventListener('scroll', onScroll);\n\n const dimensions = getWindowDimensions();\n const windowWidth = dimensions.width;\n const windowHeight = dimensions.height;\n console.log('componentDidMount() window geometry', windowWidth, windowHeight);\n\n window.addEventListener('resize', _.throttle(this.windowSizeChangeHandle, 500));\n globalSettings.maxCodePlaySpeed = getUxSettings().MAX_CODE_PLAY_SPEED;\n\n this.setState({\n windowWidth,\n windowHeight,\n mounted: true,\n });\n win.setAll(windowWidth, windowHeight, window.scrollY, true);\n }\n\n componentWillUnmount() {\n window.removeEventListener('resize', this.windowSizeChangeHandle);\n }\n\n render() {\n console.log('App.render()');\n const contents = ;\n const independentContents = this.props.selectedChapterId === 'chapter1';\n // Make sure SSR works\n const {windowWidth, windowHeight} = this.state.mounted ? this.state : {};\n\n let chapters = [];\n for (let [i, Chapter] of this.props.chapters.entries()) {\n chapters.push(\n \n \n \n );\n }\n return (\n \n
\n \n \n \n

Inside python dict — an explorable explanation

\n \n \n \n {!independentContents && {contents}}\n {chapters}\n \n \n \n
\n