Skip to content

Jaime Collado / textflow

Go to a project
Commit be5dfe7d by Estrella Vallecillo
Options

getItem test fixed

parent e146fd66
Inline Side-by-side
Showing with 201 additions and 26 deletions
poetry.lock
......@@ -21,14 +21,33 @@ tests = ["coverage[toml] (>=5.0.2)", "hypothesis", "pympler", "pytest (>=4.3.0)"
tests_no_zope = ["coverage[toml] (>=5.0.2)", "hypothesis", "pympler", "pytest (>=4.3.0)", "six", "mypy", "pytest-mypy-plugins", "cloudpickle"]
[[package]]
name = "click"
version = "8.1.2"
description = "Composable command line interface toolkit"
category = "main"
optional = false
python-versions = ">=3.7"
[package.dependencies]
colorama = {version = "*", markers = "platform_system == \"Windows\""}
[[package]]
name = "colorama"
version = "0.4.4"
description = "Cross-platform colored terminal text."
category = "dev"
category = "main"
optional = false
python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*, !=3.4.*"
[[package]]
name = "joblib"
version = "1.1.0"
description = "Lightweight pipelining with Python functions"
category = "main"
optional = false
python-versions = ">=3.6"
[[package]]
name = "more-itertools"
version = "8.12.0"
description = "More routines for operating on iterables, beyond itertools"
......@@ -37,6 +56,28 @@ optional = false
python-versions = ">=3.5"
[[package]]
name = "nltk"
version = "3.7"
description = "Natural Language Toolkit"
category = "main"
optional = false
python-versions = ">=3.7"
[package.dependencies]
click = "*"
joblib = "*"
regex = ">=2021.8.3"
tqdm = "*"
[package.extras]
all = ["numpy", "pyparsing", "scipy", "matplotlib", "twython", "requests", "scikit-learn", "python-crfsuite"]
corenlp = ["requests"]
machine_learning = ["numpy", "python-crfsuite", "scikit-learn", "scipy"]
plot = ["matplotlib"]
tgrep = ["pyparsing"]
twitter = ["twython"]
[[package]]
name = "packaging"
version = "21.3"
description = "Core utilities for Python packages"
......@@ -100,6 +141,31 @@ checkqa-mypy = ["mypy (==v0.761)"]
testing = ["argcomplete", "hypothesis (>=3.56)", "mock", "nose", "requests", "xmlschema"]
[[package]]
name = "regex"
version = "2022.4.24"
description = "Alternative regular expression module, to replace re."
category = "main"
optional = false
python-versions = ">=3.6"
[[package]]
name = "tqdm"
version = "4.64.0"
description = "Fast, Extensible Progress Meter"
category = "main"
optional = false
python-versions = "!=3.0.*,!=3.1.*,!=3.2.*,!=3.3.*,>=2.7"
[package.dependencies]
colorama = {version = "*", markers = "platform_system == \"Windows\""}
[package.extras]
dev = ["py-make (>=0.1.0)", "twine", "wheel"]
notebook = ["ipywidgets (>=6)"]
slack = ["slack-sdk"]
telegram = ["requests"]
[[package]]
name = "wcwidth"
version = "0.2.5"
description = "Measures the displayed width of unicode strings in a terminal"
......@@ -110,7 +176,7 @@ python-versions = "*"
[metadata]
lock-version = "1.1"
python-versions = "^3.8"
content-hash = "c27944f25b55067b06883f1cea204be7d97841a4b8228fab69b91895347494ad"
content-hash = "a1cdfe07ff158499c3b646d7d232739f2f518b2f0cf3c30314d9b5d1c37e7a33"
[metadata.files]
atomicwrites = [
......@@ -121,14 +187,26 @@ attrs = [
{file = "attrs-21.4.0-py2.py3-none-any.whl", hash = "sha256:2d27e3784d7a565d36ab851fe94887c5eccd6a463168875832a1be79c82828b4"},
{file = "attrs-21.4.0.tar.gz", hash = "sha256:626ba8234211db98e869df76230a137c4c40a12d72445c45d5f5b716f076e2fd"},
]
click = [
{file = "click-8.1.2-py3-none-any.whl", hash = "sha256:24e1a4a9ec5bf6299411369b208c1df2188d9eb8d916302fe6bf03faed227f1e"},
{file = "click-8.1.2.tar.gz", hash = "sha256:479707fe14d9ec9a0757618b7a100a0ae4c4e236fac5b7f80ca68028141a1a72"},
]
colorama = [
{file = "colorama-0.4.4-py2.py3-none-any.whl", hash = "sha256:9f47eda37229f68eee03b24b9748937c7dc3868f906e8ba69fbcbdd3bc5dc3e2"},
{file = "colorama-0.4.4.tar.gz", hash = "sha256:5941b2b48a20143d2267e95b1c2a7603ce057ee39fd88e7329b0c292aa16869b"},
]
joblib = [
{file = "joblib-1.1.0-py2.py3-none-any.whl", hash = "sha256:f21f109b3c7ff9d95f8387f752d0d9c34a02aa2f7060c2135f465da0e5160ff6"},
{file = "joblib-1.1.0.tar.gz", hash = "sha256:4158fcecd13733f8be669be0683b96ebdbbd38d23559f54dca7205aea1bf1e35"},
]
more-itertools = [
{file = "more-itertools-8.12.0.tar.gz", hash = "sha256:7dc6ad46f05f545f900dd59e8dfb4e84a4827b97b3cfecb175ea0c7d247f6064"},
{file = "more_itertools-8.12.0-py3-none-any.whl", hash = "sha256:43e6dd9942dffd72661a2c4ef383ad7da1e6a3e968a927ad7a6083ab410a688b"},
]
nltk = [
{file = "nltk-3.7-py3-none-any.whl", hash = "sha256:ba3de02490308b248f9b94c8bc1ac0683e9aa2ec49ee78536d8667afb5e3eec8"},
{file = "nltk-3.7.zip", hash = "sha256:d6507d6460cec76d70afea4242a226a7542f85c669177b9c7f562b7cf1b05502"},
]
packaging = [
{file = "packaging-21.3-py3-none-any.whl", hash = "sha256:ef103e05f519cdc783ae24ea4e2e0f508a9c99b2d4969652eed6a2e1ea5bd522"},
{file = "packaging-21.3.tar.gz", hash = "sha256:dd47c42927d89ab911e606518907cc2d3a1f38bbd026385970643f9c5b8ecfeb"},
......@@ -149,6 +227,86 @@ pytest = [
{file = "pytest-5.4.3-py3-none-any.whl", hash = "sha256:5c0db86b698e8f170ba4582a492248919255fcd4c79b1ee64ace34301fb589a1"},
{file = "pytest-5.4.3.tar.gz", hash = "sha256:7979331bfcba207414f5e1263b5a0f8f521d0f457318836a7355531ed1a4c7d8"},
]
regex = [
{file = "regex-2022.4.24-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:f86aef546add4ff1202e1f31e9bb54f9268f17d996b2428877283146bf9bc013"},
{file = "regex-2022.4.24-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:e944268445b5694f5d41292c9228f0ca46d5a32a67f195d5f8547c1f1d91f4bc"},
{file = "regex-2022.4.24-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:0f8da3145f4b72f7ce6181c804eaa44cdcea313c8998cdade3d9e20a8717a9cb"},
{file = "regex-2022.4.24-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:0fd464e547dbabf4652ca5fe9d88d75ec30182981e737c07b3410235a44b9939"},
{file = "regex-2022.4.24-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:071bcb625e890f28b7c4573124a6512ea65107152b1d3ca101ce33a52dad4593"},
{file = "regex-2022.4.24-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:1c2de7f32fa87d04d40f54bce3843af430697aba51c3a114aa62837a0772f219"},
{file = "regex-2022.4.24-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:1a07e8366115069f26822c47732122ab61598830a69f5629a37ea8881487c107"},
{file = "regex-2022.4.24-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:036d1c1fbe69eba3ee253c107e71749cdbb4776db93d674bc0d5e28f30300734"},
{file = "regex-2022.4.24-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:af1e687ffab18a75409e5e5d6215b6ccd41a5a1a0ea6ce9665e01253f737a0d3"},
{file = "regex-2022.4.24-cp310-cp310-musllinux_1_1_i686.whl", hash = "sha256:165cc75cfa5aa0f12adb2ac6286330e7229a06dc0e6c004ec35da682b5b89579"},
{file = "regex-2022.4.24-cp310-cp310-musllinux_1_1_ppc64le.whl", hash = "sha256:3e35c50b27f36176c792738cb9b858523053bc495044d2c2b44db24376b266f1"},
{file = "regex-2022.4.24-cp310-cp310-musllinux_1_1_s390x.whl", hash = "sha256:43ee0df35925ae4b0cc6ee3f60b73369e559dd2ac40945044da9394dd9d3a51d"},
{file = "regex-2022.4.24-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:58521abdab76583bd41ef47e5e2ddd93b32501aee4ee8cee71dee10a45ba46b1"},
{file = "regex-2022.4.24-cp310-cp310-win32.whl", hash = "sha256:275afc7352982ee947fc88f67a034b52c78395977b5fc7c9be15f7dc95b76f06"},
{file = "regex-2022.4.24-cp310-cp310-win_amd64.whl", hash = "sha256:253f858a0255cd91a0424a4b15c2eedb12f20274f85731b0d861c8137e843065"},
{file = "regex-2022.4.24-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:85b7ee4d0c7a46296d884f6b489af8b960c4291d76aea4b22fd4fbe05e6ec08e"},
{file = "regex-2022.4.24-cp36-cp36m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8e0da7ef160d4f3eb3d4d3e39a02c3c42f7dbcfce62c81f784cc99fc7059765f"},
{file = "regex-2022.4.24-cp36-cp36m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:4f2e2cef324ca9355049ee1e712f68e2e92716eba24275e6767b9bfa15f1f478"},
{file = "regex-2022.4.24-cp36-cp36m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:6165e737acb3bea3271372e8aa5ebe7226c8a8e8da1b94af2d6547c5a09d689d"},
{file = "regex-2022.4.24-cp36-cp36m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:3f6bd8178cce5bb56336722d5569d19c50bba5915a69a2050c497fb921e7cb0f"},
{file = "regex-2022.4.24-cp36-cp36m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:45b761406777a681db0c24686178532134c937d24448d9e085279b69e9eb7da4"},
{file = "regex-2022.4.24-cp36-cp36m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:3dfbadb7b74d95f72f9f9dbf9778f7de92722ab520a109ceaf7927461fa85b10"},
{file = "regex-2022.4.24-cp36-cp36m-musllinux_1_1_aarch64.whl", hash = "sha256:9913bcf730eb6e9b441fb176832eea9acbebab6035542c7c89d90c803f5cd3be"},
{file = "regex-2022.4.24-cp36-cp36m-musllinux_1_1_i686.whl", hash = "sha256:68aed3fb0c61296bd6d234f558f78c51671f79ccb069cbcd428c2eea6fee7a5b"},
{file = "regex-2022.4.24-cp36-cp36m-musllinux_1_1_ppc64le.whl", hash = "sha256:8e7d33f93cdd01868327d834d0f5bb029241cd293b47d51b96814dec27fc9b4b"},
{file = "regex-2022.4.24-cp36-cp36m-musllinux_1_1_s390x.whl", hash = "sha256:82b7fc67e49fdce671bdbec1127189fc979badf062ce6e79dc95ef5e07a8bf92"},
{file = "regex-2022.4.24-cp36-cp36m-musllinux_1_1_x86_64.whl", hash = "sha256:c36906a7855ec33a9083608e6cd595e4729dab18aeb9aad0dd0b039240266239"},
{file = "regex-2022.4.24-cp36-cp36m-win32.whl", hash = "sha256:b2df3ede85d778c949d9bd2a50237072cee3df0a423c91f5514f78f8035bde87"},
{file = "regex-2022.4.24-cp36-cp36m-win_amd64.whl", hash = "sha256:dffd9114ade73137ab2b79a8faf864683dbd2dbbb6b23a305fbbd4cbaeeb2187"},
{file = "regex-2022.4.24-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:6a0ef57cccd8089b4249eebad95065390e56c04d4a92c51316eab4131bca96a9"},
{file = "regex-2022.4.24-cp37-cp37m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:12af15b6edb00e425f713160cfd361126e624ec0de86e74f7cad4b97b7f169b3"},
{file = "regex-2022.4.24-cp37-cp37m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:7f271d0831d8ebc56e17b37f9fa1824b0379221d1238ae77c18a6e8c47f1fdce"},
{file = "regex-2022.4.24-cp37-cp37m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:37903d5ca11fa47577e8952d2e2c6de28553b11c70defee827afb941ab2c6729"},
{file = "regex-2022.4.24-cp37-cp37m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:8b747cef8e5dcdaf394192d43a0c02f5825aeb0ecd3d43e63ae500332ab830b0"},
{file = "regex-2022.4.24-cp37-cp37m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:582ea06079a03750b5f71e20a87cd99e646d796638b5894ff85987ebf5e04924"},
{file = "regex-2022.4.24-cp37-cp37m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:aa6daa189db9104787ff1fd7a7623ce017077aa59eaac609d0d25ba95ed251a0"},
{file = "regex-2022.4.24-cp37-cp37m-musllinux_1_1_aarch64.whl", hash = "sha256:7dbc96419ef0fb6ac56626014e6d3a345aeb8b17a3df8830235a88626ffc8d84"},
{file = "regex-2022.4.24-cp37-cp37m-musllinux_1_1_i686.whl", hash = "sha256:0fb6cb16518ac7eff29d1e0b0cce90275dfae0f17154165491058c31d58bdd1d"},
{file = "regex-2022.4.24-cp37-cp37m-musllinux_1_1_ppc64le.whl", hash = "sha256:bea61de0c688198e3d9479344228c7accaa22a78b58ec408e41750ebafee6c08"},
{file = "regex-2022.4.24-cp37-cp37m-musllinux_1_1_s390x.whl", hash = "sha256:46cbc5b23f85e94161b093dba1b49035697cf44c7db3c930adabfc0e6d861b95"},
{file = "regex-2022.4.24-cp37-cp37m-musllinux_1_1_x86_64.whl", hash = "sha256:50b77622016f03989cd06ecf6b602c7a6b4ed2e3ce04133876b041d109c934ee"},
{file = "regex-2022.4.24-cp37-cp37m-win32.whl", hash = "sha256:2bde99f2cdfd6db1ec7e02d68cadd384ffe7413831373ea7cc68c5415a0cb577"},
{file = "regex-2022.4.24-cp37-cp37m-win_amd64.whl", hash = "sha256:66fb765b2173d90389384708e3e1d3e4be1148bd8d4d50476b1469da5a2f0229"},
{file = "regex-2022.4.24-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:709396c0c95b95045fac89b94f997410ff39b81a09863fe21002f390d48cc7d3"},
{file = "regex-2022.4.24-cp38-cp38-macosx_11_0_arm64.whl", hash = "sha256:7a608022f4593fc67518c6c599ae5abdb03bb8acd75993c82cd7a4c8100eff81"},
{file = "regex-2022.4.24-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:fb7107faf0168de087f62a2f2ed00f9e9da12e0b801582b516ddac236b871cda"},
{file = "regex-2022.4.24-cp38-cp38-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:aabc28f7599f781ddaeac168d0b566d0db82182cc3dcf62129f0a4fc2927b811"},
{file = "regex-2022.4.24-cp38-cp38-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:92ad03f928675ca05b79d3b1d3dfc149e2226d57ed9d57808f82105d511d0212"},
{file = "regex-2022.4.24-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:b7ba3c304a4a5d8112dbd30df8b3e4ef59b4b07807957d3c410d9713abaee9a8"},
{file = "regex-2022.4.24-cp38-cp38-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:e2acf5c66fbb62b5fe4c40978ddebafa50818f00bf79d60569d9762f6356336e"},
{file = "regex-2022.4.24-cp38-cp38-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:7c4d9770e579eb11b582b2e2fd19fa204a15cb1589ae73cd4dcbb63b64f3e828"},
{file = "regex-2022.4.24-cp38-cp38-musllinux_1_1_aarch64.whl", hash = "sha256:02543d6d5c32d361b7cc468079ba4cddaaf4a6544f655901ba1ff9d8e3f18755"},
{file = "regex-2022.4.24-cp38-cp38-musllinux_1_1_i686.whl", hash = "sha256:73ed1b06abadbf6b61f6033a07c06f36ec0ddca117e41ef2ac37056705e46458"},
{file = "regex-2022.4.24-cp38-cp38-musllinux_1_1_ppc64le.whl", hash = "sha256:3241db067a7f69da57fba8bca543ac8a7ca415d91e77315690202749b9fdaba1"},
{file = "regex-2022.4.24-cp38-cp38-musllinux_1_1_s390x.whl", hash = "sha256:d128e278e5e554c5c022c7bed410ca851e00bacebbb4460de546a73bc53f8de4"},
{file = "regex-2022.4.24-cp38-cp38-musllinux_1_1_x86_64.whl", hash = "sha256:b1d53835922cd0f9b74b2742453a444865a70abae38d12eb41c59271da66f38d"},
{file = "regex-2022.4.24-cp38-cp38-win32.whl", hash = "sha256:f2a5d9f612091812dee18375a45d046526452142e7b78c4e21ab192db15453d5"},
{file = "regex-2022.4.24-cp38-cp38-win_amd64.whl", hash = "sha256:a850f5f369f1e3b6239da7fb43d1d029c1e178263df671819889c47caf7e4ff3"},
{file = "regex-2022.4.24-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:bedb3d01ad35ea1745bdb1d57f3ee0f996f988c98f5bbae9d068c3bb3065d210"},
{file = "regex-2022.4.24-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:8bf867ba71856414a482e4b683500f946c300c4896e472e51d3db8dfa8dc8f32"},
{file = "regex-2022.4.24-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:b415b82e5be7389ec5ee7ee35431e4a549ea327caacf73b697c6b3538cb5c87f"},
{file = "regex-2022.4.24-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:9dae5affbb66178dad6c6fd5b02221ca9917e016c75ee3945e9a9563eb1fbb6f"},
{file = "regex-2022.4.24-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:e65580ae3137bce712f505ec7c2d700aef0014a3878c4767b74aff5895fc454f"},
{file = "regex-2022.4.24-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:3e9e983fc8e0d4d5ded7caa5aed39ca2cf6026d7e39801ef6f0af0b1b6cd9276"},
{file = "regex-2022.4.24-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:cfad3a770839aa456ff9a9aa0e253d98b628d005a3ccb37da1ff9be7c84fee16"},
{file = "regex-2022.4.24-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:ed625205f5f26984382b68e4cbcbc08e6603c9e84c14b38457170b0cc71c823b"},
{file = "regex-2022.4.24-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:c4fdf837666f7793a5c3cfa2f2f39f03eb6c7e92e831bc64486c2f547580c2b3"},
{file = "regex-2022.4.24-cp39-cp39-musllinux_1_1_i686.whl", hash = "sha256:ed26c3d2d62c6588e0dad175b8d8cc0942a638f32d07b80f92043e5d73b7db67"},
{file = "regex-2022.4.24-cp39-cp39-musllinux_1_1_ppc64le.whl", hash = "sha256:f89d26e50a4c7453cb8c415acd09e72fbade2610606a9c500a1e48c43210a42d"},
{file = "regex-2022.4.24-cp39-cp39-musllinux_1_1_s390x.whl", hash = "sha256:97af238389cb029d63d5f2d931a7e8f5954ad96e812de5faaed373b68e74df86"},
{file = "regex-2022.4.24-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:be392d9cd5309509175a9d7660dc17bf57084501108dbff0c5a8bfc3646048c3"},
{file = "regex-2022.4.24-cp39-cp39-win32.whl", hash = "sha256:bcc6f7a3a95119c3568c572ca167ada75f8319890706283b9ba59b3489c9bcb3"},
{file = "regex-2022.4.24-cp39-cp39-win_amd64.whl", hash = "sha256:5b9c7b6895a01204296e9523b3e12b43e013835a9de035a783907c2c1bc447f0"},
{file = "regex-2022.4.24.tar.gz", hash = "sha256:92183e9180c392371079262879c6532ccf55f808e6900df5d9f03c9ca8807255"},
]
tqdm = [
{file = "tqdm-4.64.0-py2.py3-none-any.whl", hash = "sha256:74a2cdefe14d11442cedf3ba4e21a3b84ff9a2dbdc6cfae2c34addb2a14a5ea6"},
{file = "tqdm-4.64.0.tar.gz", hash = "sha256:40be55d30e200777a307a7585aee69e4eabb46b4ec6a4b4a5f2d9f11e7d5408d"},
]
wcwidth = [
{file = "wcwidth-0.2.5-py2.py3-none-any.whl", hash = "sha256:beb4802a9cebb9144e99086eff703a642a13d6a0052920003a230f3294bbe784"},
{file = "wcwidth-0.2.5.tar.gz", hash = "sha256:c4d647b99872929fdb7bdcaa4fbe7f01413ed3d98077df798530e5b04f116c83"},
......
pyproject.toml
......@@ -6,6 +6,7 @@ authors = ["Jaime Collado <jcollado@ujaen.es>", "Estrella Vallecillo <mevr0003@r
[tool.poetry.dependencies]
python = "^3.8"
nltk = "^3.7"
[tool.poetry.dev-dependencies]
pytest = "^5.2"
......
tests/test_sequence.py
......@@ -83,45 +83,42 @@ def test_len(sequence, expected):
[
pytest.param(
Sequence("string", "Lorem ipsum"),
[Sequence() for _ in range(2)]
[Sequence("token",tkn) for tkn in "Lorem ipsum".split(" ")]
),
pytest.param(
Sequence("text", "tests/data/doc_1.txt"),
{
"child": [("string", "Lorem ipsum dolor sit amet"), ("string", "Nam lectus turpis")],
"sequence": [Sequence() for _ in range(2)]
}
[Sequence("token",tkn) for tkn in "Lorem ipsum".split(" ")]
),
pytest.param(
Sequence("directory","tests/data" ),
2
[Sequence().initFromDocument("tests/data/doc_1.txt","tokens","token")]
)
]
)
def test_iter(sequence, expected):
assert iter(sequence).__next__() == expected
@pytest.mark.parametrize(
"sequence, expected",
[
pytest.param(
Sequence("string", "Lorem ipsum dolor sit amet"),
[Sequence() for _ in range(5)]
Sequence("string", "Lorem ipsum dolor sit amet").children["tokens"]
),
pytest.param(
Sequence("text", "tests/data/doc_1.txt"),
[Sequence() for _ in range(8)]
Sequence("text", "tests/data/doc_1.txt").children["tokens"]
),
pytest.param(
Sequence("directory","tests/data" ),
2
Sequence("directory","tests/data" ).children["files"]
)
]
)
def test_getitem(sequence, expected):
assert sequence[0] == expected
@pytest.mark.parametrize(
"sequence, expected",
[
......
textflow/Sequence.py
import os
from typing import Optional
from nltk.tokenize import TreebankWordTokenizer
from nltk.tokenize import WhitespaceTokenizer
from nltk.tokenize import SpaceTokenizer
from nltk.tokenize import WordPunctTokenizer
class SequenceIterator: #TODO documentar
class SequenceIterator:
def __init__(self, children):
"""
Creates a sequenceIterator from a Sequence.
......@@ -47,7 +51,7 @@ class Sequence:
text: ...
sequences: ...
"""
def __init__(self, format: Optional[str] = None, src: Optional[object] = None):
def __init__(self, format: Optional[str] = None, src: Optional[object] = None, tokenizer: Optional[object] = None ):
"""Creates a sequence from an input object.
Args:
......@@ -61,6 +65,8 @@ class Sequence:
raise ValueError(
f"{format} is not a valid format. Valid formats: {_VALID_FORMATS}"
)
if tokenizer == None:
tokenizer = WhitespaceTokenizer()
self.format = format
self.children = {}
......@@ -70,13 +76,13 @@ class Sequence:
raise ValueError(f"{src} is not an instance of token")
self.metadata["text"] = src
if format == "string":
self.initFromString(src,"tokens","token")
self.initFromString(src,"tokens","token",tokenizer)
if format == "text":
self.initFromDocument(src,"tokens","token")
self.initFromDocument(src,"tokens","token", tokenizer)
if format == "directory":
self.initFromDirectory(src,"directory","files")
self.initFromDirectory(src,"directory","files",tokenizer)
def initFromDirectory(self, directory, labelDirectory, labelFile ): #TODO Inicializador por defecto para un directorio
def initFromDirectory(self, directory, labelDirectory, labelFile, tokenizer):
'''
Initialize a Sequence from a directory
Args:
......@@ -101,12 +107,12 @@ class Sequence:
else:
self.metadata["directoriesPath"].append(directory+"/"+file)
if labelDirectory in self.children:
self.children[labelDirectory].append(Sequence("directory", directory+"/"+file ))
self.children[labelDirectory].append(Sequence("directory", directory+"/"+file,tokenizer ))
else:
self.children[labelDirectory]= [Sequence("directory", directory+"/"+file)]
self.children[labelDirectory]= [Sequence("directory", directory+"/"+file, tokenizer)]
def initFromDocument(self, documentPath, labelSubSequence, formatSubsequence):
def initFromDocument(self, documentPath, labelSubSequence, formatSubsequence, tokenizer):
'''
Initialize a Sequence from a document
Args:
......@@ -117,10 +123,10 @@ class Sequence:
self.format = "text"
with open(documentPath, "r") as f:
txt = f.read()
self.children[labelSubSequence] = [Sequence(formatSubsequence,token_src) for token_src in txt.split(" ")]
self.children[labelSubSequence] = [Sequence(formatSubsequence,token_src) for token_src in tokenizer.tokenize(txt)]
self.metadata["text"] = txt
def initFromString(self, srcString, labelSubSequence, formatSubsequence):
def initFromString(self, srcString, labelSubSequence, formatSubsequence, tokenizer):
'''
Initialize a Sequence from a string
Args:
......@@ -133,7 +139,7 @@ class Sequence:
if not isinstance(srcString, str):
raise ValueError(f"{srcString} is not an instance of string")
self.format = "string"
self.children[labelSubSequence]= [Sequence(formatSubsequence,token_src) for token_src in srcString.split(" ")]
self.children[labelSubSequence]= [Sequence(formatSubsequence,token_src) for token_src in tokenizer.tokenize(srcString)]
self.metadata["text"]= srcString
......@@ -180,8 +186,15 @@ class Sequence:
'''
return SequenceIterator(list(self.children.values()))
def __getitem__(self, idx): #TODO Documentacion
def __getitem__(self, idx):
'''
Get the value of a key from the dictionary of children
Args:
idx: a string that represent the key of the children dictionary
or an integer that represent the position of the key in children dictionary keys
Returns:
A List of Sequences
'''
if isinstance(idx, str): # Get src by string (e.g. seq["doc1"])
if self.children:
if idx in self.children:
......@@ -198,6 +211,12 @@ class Sequence:
else: # TODO: Should it support slices (e.g. [2:4])?
raise ValueError(f"Sequence id '{idx}' not found in {self.children}")
def __eq__(self, other):
if self.format == other.format and self.metadata == other.metadata and self.children == other.children:
return True
else:
return False
def depth(self,diccionaryList: Optional[list] = None):
'''
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or sign in to comment