=============
``py3langid``
=============
``py3langid`` is a fork of the standalone language identification tool ``langid.py`` by Marco Lui.
Original license: BSD-2-Clause. Fork license: BSD-3-Clause.
Changes in this fork
--------------------
Execution speed has been improved and the code base has been optimized for Python 3.6+:
- Import: Loading the package (``import py3langid``) is about 30% faster
- Startup: Loading the default classification model is 25-30x faster
- Execution: Language detection with ``langid.classify`` is 5-6x faster on paragraphs (less on longer texts)
For implementation details see this blog post: `How to make language detection with langid.py faster <https://adrien.barbaresi.eu/blog/language-detection-langid-py-faster.html>`_.
For more information and older Python versions see `changelog <https://github.com/adbar/py3langid/blob/master/HISTORY.rst>`_.
Usage
-----
Drop-in replacement
~~~~~~~~~~~~~~~~~~~
1. Install the package:
* ``pip3 install py3langid`` (or ``pip`` where applicable)
2. Use it:
* with Python: ``import py3langid as langid``
* on the command-line: ``langid``
With Python
~~~~~~~~~~~
Basics:
.. code-block:: python
>>> import py3langid as langid
>>> text = 'This text is in English.'
# identified language and probability
>>> langid.classify(text)
('en', -56.77429)
# unpack the result tuple in variables
>>> lang, prob = langid.classify(text)
# all potential languages
>>> langid.rank(text)
More options:
.. code-block:: python
>>> from py3langid.langid import LanguageIdentifier, MODEL_FILE
# subset of target languages
>>> identifier = LanguageIdentifier.from_pickled_model(MODEL_FILE)
>>> identifier.set_languages(['de', 'en', 'fr'])
# this won't work well...
>>> identifier.classify('这样不好')
('en', -81.831665)
# normalization of probabilities to an interval between 0 and 1
>>> identifier = LanguageIdentifier.from_pickled_model(MODEL_FILE, norm_probs=True)
>>> identifier.classify('This should be enough text.')
('en', 1.0)
Note: the Numpy data type for the feature vector has been changed to optimize for speed. If results are inconsistent, try restoring the original setting:
.. code-block:: python
>>> langid.classify(text, datatype='uint32')
On the command-line
~~~~~~~~~~~~~~~~~~~
.. code-block:: bash
# basic usage with probability normalization
$ echo "This should be enough text." | langid -n
('en', 1.0)
# define a subset of target languages
$ echo "This won't be recognized properly." | langid -n -l fr,it,tr
('it', 0.97038305)
Legacy documentation
--------------------
**The docs below are provided for reference, only part of the functions are currently tested and maintained.**
Introduction
------------
``langid.py`` is a standalone Language Identification (LangID) tool.
The design principles are as follows:
1. Fast
2. Pre-trained over a large number of languages (currently 97)
3. Not sensitive to domain-specific features (e.g. HTML/XML markup)
4. Single .py file with minimal dependencies
5. Deployable as a web service
All that is required to run ``langid.py`` is Python >= 3.6 and numpy.
The accompanying training tools are still Python2-only.
``langid.py`` is WSGI-compliant. ``langid.py`` will use ``fapws3`` as a web server if
available, and default to ``wsgiref.simple_server`` otherwise.
``langid.py`` comes pre-trained on 97 languages (ISO 639-1 codes given):
af, am, an, ar, as, az, be, bg, bn, br,
bs, ca, cs, cy, da, de, dz, el, en, eo,
es, et, eu, fa, fi, fo, fr, ga, gl, gu,
he, hi, hr, ht, hu, hy, id, is, it, ja,
jv, ka, kk, km, kn, ko, ku, ky, la, lb,
lo, lt, lv, mg, mk, ml, mn, mr, ms, mt,
nb, ne, nl, nn, no, oc, or, pa, pl, ps,
pt, qu, ro, ru, rw, se, si, sk, sl, sq,
sr, sv, sw, ta, te, th, tl, tr, ug, uk,
ur, vi, vo, wa, xh, zh, zu
The training data was drawn from 5 different sources:
* JRC-Acquis
* ClueWeb 09
* Wikipedia
* Reuters RCV2
* Debian i18n
Usage
-----
langid [options]
optional arguments:
-h, --help show this help message and exit
-s, --serve launch web service
--host=HOST host/ip to bind to
--port=PORT port to listen on
-v increase verbosity (repeat for greater effect)
-m MODEL load model from file
-l LANGS, --langs=LANGS
comma-separated set of target ISO639 language codes
(e.g en,de)
-r, --remote auto-detect IP address for remote access
-b, --batch specify a list of files on the command line
-d, --dist show full distribution over languages
-u URL, --url=URL langid of URL
--line process pipes line-by-line rather than as a document
-n, --normalize normalize confidence scores to probability values
The simplest way to use ``langid.py`` is as a command-line tool, and you can
invoke using ``python langid.py``. If you installed ``langid.py`` as a Python
module (e.g. via ``pip install langid``), you can invoke ``langid`` instead of
``python langid.py -n`` (the two are equivalent). This will cause a prompt to
display. Enter text to identify, and hit enter::
>>> This is a test
('en', -54.41310358047485)
>>> Questa e una prova
('it', -35.41771221160889)
``langid.py`` can also detect when the input is redirected (only tested under Linux), and in this
case will process until EOF rather than until newline like in interactive mode::
python langid.py < README.rst
('en', -22552.496054649353)
The value returned is the unnormalized probability estimate for the language. Calculating
the exact probability estimate is disabled by default, but can be enabled through a flag::
python langid.py -n < README.rst
('en', 1.0)
More details are provided in this README in the section on `Probability Normalization`.
You can also use ``langid.py`` as a Python library::
# python
Python 2.7.2+ (default, Oct 4 2011, 20:06:09)
[GCC 4.6.1] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> import langid
>>> langid.classify("This is a test")
('en', -54.41310358047485)
Finally, ``langid.py`` can use Python's built-in ``wsgiref.simple_server`` (or ``fapws3`` if available) to
provide language identification as a web service. To do this, launch ``python langid.py -s``, and
access http://localhost:9008/detect . The web service supports GET, POST and PUT. If GET is performed
with no data, a simple HTML forms interface is displayed.
The response is generated in JSON, here is an example::
{"responseData": {"confidence": -54.41310358047485, "language": "en"}, "responseDetails": null, "responseStatus": 200}
A utility such as curl can be used to access the web service::
# curl -d "q=This is a test" localhost:9008/detect
{"responseData": {"confidence": -54.41310358047485, "language": "en"}, "responseDetails": null, "responseStatus": 200}
You can also use HTTP PUT::
# curl -T readme.rst localhost:9008/detect
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 2871 100 119 100 2752 117 2723 0:00:01 0:00:01 --:--:-- 2727
{"responseData": {"confidence": -22552.496054649353, "language": "en"}, "responseDetails": null, "responseStatus": 200}
If no "q=XXX" key-value pair is present in the HTTP POST payload, ``langid.py`` will interpret the entire
file as a single query. This allows for redirection via curl::
# echo "This is a test" | curl -d @- localhost:9008/detect
{"responseData": {"confidence": -54.41310358047485, "language": "en"}, "responseDetails": null, "responseStatus": 200}
``langid.py`` will attempt to discover the host IP address automatically. Often, this is set to localhost(127.0.1.1), even
though the machine has a different external IP address. ``langid.py`` can attempt to automatically discover the external
IP address. To enable this functionality, start ``langid.py`` with the ``-r`` flag.
``langid.py`` supports constraining of the output language set using the ``-l`` flag and a comma-separated list of ISO639-1
language codes (the ``-n`` flag enables probability normalization)::
# python langid.py -n -l it,fr
>>> Io non parlo italiano
('it', 0.99999999988965627)
>>> Je ne parle pas français
('fr', 1.0)
>>> I don't speak english
('it', 0.92210605672341062)
When using ``langid.py`` as a library, the set_languages method can be used to constrain the language set::
python
Python 2.7.2+ (default, Oct 4 2011, 20:06:09)
[GCC 4.6.1] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> import langid
>>> langid.classify("I do not speak english")
('en', 0.57133487679900674)
>>> langid.set_languages(['de','fr','it'])
>>> langid.classify("I do not speak english")
('it', 0.99999835791478453)
>>> langid.set_languages(['en','it'])
>>> langid.classify("I do not speak english")
('en', 0.99176190378750373)
Batch Mode
----------
``langid.py`` supports batch mode processing, which can be invoked with the ``-b`` flag.
In this mode, ``langid.py`` reads a list of paths to files to classify as arguments.
If no arguments are supplied, ``langid.py`` reads the list of paths from ``stdin``,
this is useful for using ``langid.py`` with UNIX utilities such as ``find``.
In batch mode, ``langid.py`` uses ``multiprocessing`` to invoke multiple instances of
the classifier, utilizing all available CPUs to classify documents in parallel.
Probability Normalization
-------------------------
The probabilistic model implemented by ``langid.py`` involves the multiplication of a
large number of probabilities. For computational reasons, the actual calculations are
implemented in the log-probability space (a common numerical technique for dealing with
vanishingly small probabilities). One side-effect of this is that it is not necessary to
compute a full probability in order to determine the most probable language in a set
of candidate languages. However, users sometimes find it helpful to have a "confidence"
score for the probability prediction. Thus, ``langid.py`` implements a re-normalization
that produces an output in the 0-1 range.
``langid.py`` disables probability normalization by default. For
command-line usages of ``langid.py``, it can be enabled by passing the ``-n`` flag. For
probability normalization in library use, the user must instantiate their own
``LanguageIdentifier``. An example of such usage is as follows::
>> from py3langid.langid import LanguageIdentifier, MODEL_FILE
>> identifier = LanguageIdentifier.from_pickled_model(MODEL_FILE, norm_probs=True)
>> identifier.classify("This is a test")
('en', 0.9999999909903544)
Training a model
----------------
So far Python 2.7 only, see the `original instructions <https://github.com/saffsd/langid.py#training-a-model>`_.
Read more
---------
``langid.py`` is based on published research. [1] describes the LD feature selection technique in detail,
and [2] provides more detail about the module ``langid.py`` itself.
[1] Lui, Marco and Timothy Baldwin (2011) Cross-domain Feature Selection for Language Identification,
In Proceedings of the Fifth International Joint Conference on Natural Language Processing (IJCNLP 2011),
Chiang Mai, Thailand, pp. 553—561. Available from http://www.aclweb.org/anthology/I11-1062
[2] Lui, Marco and Timothy Baldwin (2012) langid.py: An Off-the-shelf Language Identification Tool,
In Proceedings of the 50th Annual Meeting of the Association for Computational Linguistics (ACL 2012),
Demo Session, Jeju, Republic of Korea. Available from www.aclweb.org/anthology/P12-3005
Raw data
{
"_id": null,
"home_page": null,
"name": "py3langid",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.8",
"maintainer_email": null,
"keywords": "language detection, language identification, langid, langid.py",
"author": "Marco Lui",
"author_email": "Adrien Barbaresi <barbaresi@bbaw.de>",
"download_url": "https://files.pythonhosted.org/packages/99/43/c3f7a3c5150c56a0ca70c3039e53cc58046698b7ce0913bb8fa86d71abcb/py3langid-0.3.0.tar.gz",
"platform": null,
"description": "=============\n``py3langid``\n=============\n\n\n``py3langid`` is a fork of the standalone language identification tool ``langid.py`` by Marco Lui.\n\nOriginal license: BSD-2-Clause. Fork license: BSD-3-Clause.\n\n\n\nChanges in this fork\n--------------------\n\nExecution speed has been improved and the code base has been optimized for Python 3.6+:\n\n- Import: Loading the package (``import py3langid``) is about 30% faster\n- Startup: Loading the default classification model is 25-30x faster\n- Execution: Language detection with ``langid.classify`` is 5-6x faster on paragraphs (less on longer texts)\n\nFor implementation details see this blog post: `How to make language detection with langid.py faster <https://adrien.barbaresi.eu/blog/language-detection-langid-py-faster.html>`_.\n\nFor more information and older Python versions see `changelog <https://github.com/adbar/py3langid/blob/master/HISTORY.rst>`_.\n\n\nUsage\n-----\n\nDrop-in replacement\n~~~~~~~~~~~~~~~~~~~\n\n\n1. Install the package:\n\n * ``pip3 install py3langid`` (or ``pip`` where applicable)\n\n2. Use it:\n\n * with Python: ``import py3langid as langid``\n * on the command-line: ``langid``\n\n\nWith Python\n~~~~~~~~~~~\n\nBasics:\n\n.. code-block:: python\n\n >>> import py3langid as langid\n \n >>> text = 'This text is in English.'\n # identified language and probability\n >>> langid.classify(text)\n ('en', -56.77429)\n # unpack the result tuple in variables\n >>> lang, prob = langid.classify(text)\n # all potential languages\n >>> langid.rank(text)\n\n\nMore options:\n\n.. code-block:: python\n\n >>> from py3langid.langid import LanguageIdentifier, MODEL_FILE\n\n # subset of target languages\n >>> identifier = LanguageIdentifier.from_pickled_model(MODEL_FILE)\n >>> identifier.set_languages(['de', 'en', 'fr'])\n # this won't work well...\n >>> identifier.classify('\u8fd9\u6837\u4e0d\u597d')\n ('en', -81.831665)\n\n # normalization of probabilities to an interval between 0 and 1\n >>> identifier = LanguageIdentifier.from_pickled_model(MODEL_FILE, norm_probs=True)\n >>> identifier.classify('This should be enough text.')\n ('en', 1.0)\n\n\nNote: the Numpy data type for the feature vector has been changed to optimize for speed. If results are inconsistent, try restoring the original setting:\n\n.. code-block:: python\n\n >>> langid.classify(text, datatype='uint32')\n\n\nOn the command-line\n~~~~~~~~~~~~~~~~~~~\n\n.. code-block:: bash\n\n # basic usage with probability normalization\n $ echo \"This should be enough text.\" | langid -n\n ('en', 1.0)\n\n # define a subset of target languages\n $ echo \"This won't be recognized properly.\" | langid -n -l fr,it,tr\n ('it', 0.97038305)\n\n\nLegacy documentation\n--------------------\n\n\n**The docs below are provided for reference, only part of the functions are currently tested and maintained.**\n\n\nIntroduction\n------------\n\n``langid.py`` is a standalone Language Identification (LangID) tool.\n\nThe design principles are as follows:\n\n1. Fast\n2. Pre-trained over a large number of languages (currently 97)\n3. Not sensitive to domain-specific features (e.g. HTML/XML markup)\n4. Single .py file with minimal dependencies\n5. Deployable as a web service\n\nAll that is required to run ``langid.py`` is Python >= 3.6 and numpy. \n\nThe accompanying training tools are still Python2-only.\n\n``langid.py`` is WSGI-compliant. ``langid.py`` will use ``fapws3`` as a web server if \navailable, and default to ``wsgiref.simple_server`` otherwise.\n\n``langid.py`` comes pre-trained on 97 languages (ISO 639-1 codes given):\n\n af, am, an, ar, as, az, be, bg, bn, br, \n bs, ca, cs, cy, da, de, dz, el, en, eo, \n es, et, eu, fa, fi, fo, fr, ga, gl, gu, \n he, hi, hr, ht, hu, hy, id, is, it, ja, \n jv, ka, kk, km, kn, ko, ku, ky, la, lb, \n lo, lt, lv, mg, mk, ml, mn, mr, ms, mt, \n nb, ne, nl, nn, no, oc, or, pa, pl, ps, \n pt, qu, ro, ru, rw, se, si, sk, sl, sq, \n sr, sv, sw, ta, te, th, tl, tr, ug, uk, \n ur, vi, vo, wa, xh, zh, zu\n\nThe training data was drawn from 5 different sources:\n\n* JRC-Acquis \n* ClueWeb 09\n* Wikipedia\n* Reuters RCV2\n* Debian i18n\n\n\nUsage\n-----\n\n langid [options]\n\noptional arguments:\n -h, --help show this help message and exit\n -s, --serve launch web service\n --host=HOST host/ip to bind to\n --port=PORT port to listen on\n -v increase verbosity (repeat for greater effect)\n -m MODEL load model from file\n -l LANGS, --langs=LANGS\n comma-separated set of target ISO639 language codes\n (e.g en,de)\n -r, --remote auto-detect IP address for remote access\n -b, --batch specify a list of files on the command line\n -d, --dist show full distribution over languages\n -u URL, --url=URL langid of URL\n --line process pipes line-by-line rather than as a document\n -n, --normalize normalize confidence scores to probability values\n\n\nThe simplest way to use ``langid.py`` is as a command-line tool, and you can \ninvoke using ``python langid.py``. If you installed ``langid.py`` as a Python \nmodule (e.g. via ``pip install langid``), you can invoke ``langid`` instead of \n``python langid.py -n`` (the two are equivalent). This will cause a prompt to \ndisplay. Enter text to identify, and hit enter::\n\n >>> This is a test\n ('en', -54.41310358047485)\n >>> Questa e una prova\n ('it', -35.41771221160889)\n\n\n``langid.py`` can also detect when the input is redirected (only tested under Linux), and in this\ncase will process until EOF rather than until newline like in interactive mode::\n\n python langid.py < README.rst \n ('en', -22552.496054649353)\n\n\nThe value returned is the unnormalized probability estimate for the language. Calculating \nthe exact probability estimate is disabled by default, but can be enabled through a flag::\n\n python langid.py -n < README.rst \n ('en', 1.0)\n\nMore details are provided in this README in the section on `Probability Normalization`.\n\nYou can also use ``langid.py`` as a Python library::\n\n # python\n Python 2.7.2+ (default, Oct 4 2011, 20:06:09) \n [GCC 4.6.1] on linux2\n Type \"help\", \"copyright\", \"credits\" or \"license\" for more information.\n >>> import langid\n >>> langid.classify(\"This is a test\")\n ('en', -54.41310358047485)\n \nFinally, ``langid.py`` can use Python's built-in ``wsgiref.simple_server`` (or ``fapws3`` if available) to\nprovide language identification as a web service. To do this, launch ``python langid.py -s``, and\naccess http://localhost:9008/detect . The web service supports GET, POST and PUT. If GET is performed\nwith no data, a simple HTML forms interface is displayed.\n\nThe response is generated in JSON, here is an example::\n\n {\"responseData\": {\"confidence\": -54.41310358047485, \"language\": \"en\"}, \"responseDetails\": null, \"responseStatus\": 200}\n\nA utility such as curl can be used to access the web service::\n\n # curl -d \"q=This is a test\" localhost:9008/detect\n {\"responseData\": {\"confidence\": -54.41310358047485, \"language\": \"en\"}, \"responseDetails\": null, \"responseStatus\": 200}\n\nYou can also use HTTP PUT::\n\n # curl -T readme.rst localhost:9008/detect\n % Total % Received % Xferd Average Speed Time Time Time Current\n Dload Upload Total Spent Left Speed\n 100 2871 100 119 100 2752 117 2723 0:00:01 0:00:01 --:--:-- 2727\n {\"responseData\": {\"confidence\": -22552.496054649353, \"language\": \"en\"}, \"responseDetails\": null, \"responseStatus\": 200}\n\nIf no \"q=XXX\" key-value pair is present in the HTTP POST payload, ``langid.py`` will interpret the entire\nfile as a single query. This allows for redirection via curl::\n\n # echo \"This is a test\" | curl -d @- localhost:9008/detect\n {\"responseData\": {\"confidence\": -54.41310358047485, \"language\": \"en\"}, \"responseDetails\": null, \"responseStatus\": 200}\n\n``langid.py`` will attempt to discover the host IP address automatically. Often, this is set to localhost(127.0.1.1), even \nthough the machine has a different external IP address. ``langid.py`` can attempt to automatically discover the external\nIP address. To enable this functionality, start ``langid.py`` with the ``-r`` flag.\n\n``langid.py`` supports constraining of the output language set using the ``-l`` flag and a comma-separated list of ISO639-1 \nlanguage codes (the ``-n`` flag enables probability normalization)::\n\n # python langid.py -n -l it,fr\n >>> Io non parlo italiano\n ('it', 0.99999999988965627)\n >>> Je ne parle pas fran\u00e7ais\n ('fr', 1.0)\n >>> I don't speak english\n ('it', 0.92210605672341062)\n\nWhen using ``langid.py`` as a library, the set_languages method can be used to constrain the language set::\n\n python \n Python 2.7.2+ (default, Oct 4 2011, 20:06:09) \n [GCC 4.6.1] on linux2\n Type \"help\", \"copyright\", \"credits\" or \"license\" for more information.\n >>> import langid\n >>> langid.classify(\"I do not speak english\")\n ('en', 0.57133487679900674)\n >>> langid.set_languages(['de','fr','it'])\n >>> langid.classify(\"I do not speak english\")\n ('it', 0.99999835791478453)\n >>> langid.set_languages(['en','it'])\n >>> langid.classify(\"I do not speak english\")\n ('en', 0.99176190378750373)\n\n\nBatch Mode\n----------\n\n``langid.py`` supports batch mode processing, which can be invoked with the ``-b`` flag.\nIn this mode, ``langid.py`` reads a list of paths to files to classify as arguments.\nIf no arguments are supplied, ``langid.py`` reads the list of paths from ``stdin``,\nthis is useful for using ``langid.py`` with UNIX utilities such as ``find``.\n\nIn batch mode, ``langid.py`` uses ``multiprocessing`` to invoke multiple instances of\nthe classifier, utilizing all available CPUs to classify documents in parallel. \n\n\nProbability Normalization\n-------------------------\n\nThe probabilistic model implemented by ``langid.py`` involves the multiplication of a\nlarge number of probabilities. For computational reasons, the actual calculations are\nimplemented in the log-probability space (a common numerical technique for dealing with\nvanishingly small probabilities). One side-effect of this is that it is not necessary to\ncompute a full probability in order to determine the most probable language in a set\nof candidate languages. However, users sometimes find it helpful to have a \"confidence\"\nscore for the probability prediction. Thus, ``langid.py`` implements a re-normalization\nthat produces an output in the 0-1 range.\n\n``langid.py`` disables probability normalization by default. For\ncommand-line usages of ``langid.py``, it can be enabled by passing the ``-n`` flag. For\nprobability normalization in library use, the user must instantiate their own \n``LanguageIdentifier``. An example of such usage is as follows::\n \n >> from py3langid.langid import LanguageIdentifier, MODEL_FILE\n >> identifier = LanguageIdentifier.from_pickled_model(MODEL_FILE, norm_probs=True)\n >> identifier.classify(\"This is a test\")\n ('en', 0.9999999909903544)\n\n\nTraining a model\n----------------\n\nSo far Python 2.7 only, see the `original instructions <https://github.com/saffsd/langid.py#training-a-model>`_.\n\n\nRead more\n---------\n\n``langid.py`` is based on published research. [1] describes the LD feature selection technique in detail,\nand [2] provides more detail about the module ``langid.py`` itself.\n\n[1] Lui, Marco and Timothy Baldwin (2011) Cross-domain Feature Selection for Language Identification, \nIn Proceedings of the Fifth International Joint Conference on Natural Language Processing (IJCNLP 2011), \nChiang Mai, Thailand, pp. 553\u2014561. Available from http://www.aclweb.org/anthology/I11-1062\n\n[2] Lui, Marco and Timothy Baldwin (2012) langid.py: An Off-the-shelf Language Identification Tool, \nIn Proceedings of the 50th Annual Meeting of the Association for Computational Linguistics (ACL 2012), \nDemo Session, Jeju, Republic of Korea. Available from www.aclweb.org/anthology/P12-3005\n",
"bugtrack_url": null,
"license": "BSD",
"summary": "Fork of the language identification tool langid.py, featuring a modernized codebase and faster execution times.",
"version": "0.3.0",
"project_urls": {
"Blog": "https://adrien.barbaresi.eu/blog/language-detection-langid-py-faster.html",
"Homepage": "https://github.com/adbar/py3langid",
"Tracker": "https://github.com/adbar/py3langid/issues"
},
"split_keywords": [
"language detection",
" language identification",
" langid",
" langid.py"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "9d1c8212ea872d236af0aea37043fb6feeaa9a43449183782b19d342f8ddd343",
"md5": "3af78872b7419e22d74a93f799a1eb84",
"sha256": "38f022eec31cf9a2bf6f142acb2a9b350fd7d0d5ae7762b1392c6d3567401fd3"
},
"downloads": -1,
"filename": "py3langid-0.3.0-py3-none-any.whl",
"has_sig": false,
"md5_digest": "3af78872b7419e22d74a93f799a1eb84",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.8",
"size": 746125,
"upload_time": "2024-06-18T11:30:51",
"upload_time_iso_8601": "2024-06-18T11:30:51.265961Z",
"url": "https://files.pythonhosted.org/packages/9d/1c/8212ea872d236af0aea37043fb6feeaa9a43449183782b19d342f8ddd343/py3langid-0.3.0-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "9943c3f7a3c5150c56a0ca70c3039e53cc58046698b7ce0913bb8fa86d71abcb",
"md5": "7e45e4e22f94a8308a115ffb58859750",
"sha256": "0a875a031a58aaf9dbda7bb8285fd75e801a7bd276216ffabe037901d4b449ec"
},
"downloads": -1,
"filename": "py3langid-0.3.0.tar.gz",
"has_sig": false,
"md5_digest": "7e45e4e22f94a8308a115ffb58859750",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.8",
"size": 752935,
"upload_time": "2024-06-18T11:31:04",
"upload_time_iso_8601": "2024-06-18T11:31:04.301853Z",
"url": "https://files.pythonhosted.org/packages/99/43/c3f7a3c5150c56a0ca70c3039e53cc58046698b7ce0913bb8fa86d71abcb/py3langid-0.3.0.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-06-18 11:31:04",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "adbar",
"github_project": "py3langid",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"lcname": "py3langid"
}
JSON
