Difference between revisions of "Kazakh and Tatar"

From Apertium
Jump to navigation Jump to search
(Directories)
 
(74 intermediate revisions by 3 users not shown)
Line 1: Line 1:
  +
{{TOCD}}
This is a language pair translating between [[Kazakh]] and [[Tatar]].
+
This is a language pair translating between [[Kazakh]] and [[Tatar]]. The pair is currently located in [https://github.com/apertium/apertium-kaz-tat trunk].
   
== General TODO ==
+
== General information ==
  +
* The [[Apertium-kaz|Kazakh transducer]] has {{#lst:apertium-kaz/stats|stems}} stems and [[apertium-kaz/stats|~{{:apertium-kaz/stats/average}}%]] coverage over random corpora
  +
* The [[Apertium-tat|Tatar transducer]] has {{#lst:apertium-tat/stats|stems}} stems and [[apertium-tat/stats|~{{:apertium-tat/stats/average}}%]] coverage over random corpora
   
  +
=== Demonstration ===
See [[/Work_plan]].
 
  +
*<code>$ echo "Бұл аударушымен қазақша жазылған мәтіндерді татаршаға аударып оқуға болады." | apertium -d . kaz-tat</code>
  +
: <code>Бу тәрҗемәче белән казакъча язылган текстларны татарчага тәрҗемә итеп укуга була.</code>
   
  +
== Installation ==
# Declination of Tatar nouns ending with -и.
 
  +
You will need:
# <s>Set up <code>bidix-with-context.sh</code> script (see <code>apertium-kaz-tat/dev/bidix</code>; seems to be very useful, requires another script from spectie)</s>.
 
  +
* [[hfst]] (svn ≥r1916)
# <s>Add some of the short wikipedia-article-like texts I have for evaluation into <code>texts</code> (should be ~200 words).</s>
 
  +
** foma
# Implement cont. class for compound/multiword nouns which already have possessive ending (<px3sp>), e.g. ''Қытай Халық Республикасы''.
 
  +
*** flex
## This continuation class should link only to CASE (but consider that some of them can have plural form: ''ишегаллары'').
 
  +
* [[Minimal_installation_from_SVN|apertium]]
# Add "ярты", "ярым" and "чирек" as numerals, but don't link them to common numerals cont. class.
 
  +
** lttoolbox (svn ≥r46087)
# (Lexical selection rule): ''сондай-ақ'' > ''шулай-ук''
 
  +
* [[CG]]
# <s>Fix roman numerals:
 
  +
** [http://icu-project.org/download/ ICU]
## add them to tat.lexc too;
 
  +
* [[Constraint-based lexical selection module|apertium-lex-tools]]
## change <code>LEXICON NUM-ROMAN</code> to something like this: <code>%<num%>%<ord%>: # ; </code>.</s>
 
  +
* apertium-kaz and apertium-tat
# Add transfer rule(s) to handle instrumental case of all parts-of-speech which are subject to substantivation, not only of nouns (this is one of the things which make testvoc results look bad)
 
# A separate cont.class for verbs which have causative forms ending with -дыр/-дер
 
# A "location-cases" cont. classes for some of the postpositions and location adverbs (e.g. "бире")
 
# '<code>Natinfl</code> cont. class in tat.lexc
 
# Fix "дыр<mod_ind>" thing (it doesn't pass bidix right now)
 
# '''Pronouns'''
 
## check cont. classes (note: if it looks like an overgeneration, and me is not sure about it, overgenerate in both lexc's)
 
## translate pronouns from kaz.lexc, add them to bidix and add equivalents into tat.lexc
 
## ^нигез/ни<prn><itg><px2pl><nom>
 
# '''Determiners'''
 
## "unify" cont. classes and tags
 
## add stems
 
# '''Adjectives'''
 
## personal clitics after adjectives are not implemented yet
 
   
  +
If you are using a Debian-based distro, the easiest way to get those dependencies is to install them with apt-get from [[User:Tino Didriksen]]'s [[Prerequisites for Debian|repository]].
=== Twol realated stuff ===
 
   
  +
== Developers ==
# <s>Current: <code>^миллион<num><subst><dat>$ --> миллионге</code> Should be: <code>^миллион<num><subst><dat>$ --> миллионға</code></s>
 
  +
Information on what remains to be done for this pair can be found at the [[/TODO]] list.
# <s>Current: <code>^сөйле<v><tv><coop><ger_past><loc>$ --> сөйлесгенде</code> Should be: <code>^сөйле<v><tv><coop><ger_past><loc>$ --> сөйлескенде</code></s>
 
# Deletion of soft sign "ь" before vowels in Tatar (see comments at the end of the <code>apertium-tat/apertium-tat.tat.twol</code> file)
 
# <s>Kazakh: <code>^ойна<v><tv><ifi><p1><pl>$ --> ойнадык</code> Should be: ''ойнадыҚ''</s>
 
# (tat) ''*аенда''
 
# *жатқандығын
 
# безнекенеме (accusative case before clitics); безнекенгәме
 
   
=== Discuss first ===
+
=== Development workflow ===
   
  +
We work on the transducers ([[apertium-kaz]] and [[apertium-tat]]) individually, and use a special process to import to the pair transducers that contain only the words found in the bidix. The following documents this process.
# There is only one formal form (<frm>) in Tatar, which can be both sg and plural. But in Kazakh there are two forms. Should I pretend as if in Tatar it *were* the same and duplicate the same form with a different tag or should I handle it in transfer?
 
# Consider ''турындагы'' - should it still be tagged as postposition?
 
# How to handle verbs with inner inflection (sometimes a Kazakh verb is translated with a multiword and vice versa. E.g. ''әуреле > башын әйләндер'')
 
----
 
   
  +
==== Adding words ====
Part-of-speech related TODO's and DONE's can be found here:
 
  +
In order to add a new word (and its translation equivalent) to the Kazakh-Tatar translator, you have to do the following:
  +
# add an entry in the bilingual dictionary — <code>apertium-kaz-tat.kaz-tat.dix</code> file in <code>apertium-kaz-tat</code> directory,
  +
# add an entry in the Kazakh monolingual dictionary — <code>apertium-kaz.kaz.lexc</code> file, which, as the name indicates, is in the <code>apertium-kaz</code> directory,
  +
# run <code>make</code> in <code>apertium-kaz</code>
  +
# add an entry in the Tatar monolingual dictionary — <code>apertium-tat.tat.lexc</code> file in <code>apertium-tat</code>,
  +
# run <code>make</code> in <code>apertium-tat</code>
  +
# <code>cd</code> to <code>apertium-kaz-tat</code> and run <code>make</code>.
   
  +
You have to have configured Kazakh-Tatar translator with the <code>--with-lang1</code> and <code>--with-lang2</code> options for the last step to work (see [[Minimal installation from SVN#For language pairs that depend on monolingual packages (apertium-XYZ)|here]] for more details on this). It will fetch changed files automatically, trim them and compile them.
* [[/Postadvebs|/Postadverbs]]
 
* [[/Postpositions]]
 
   
  +
There is '''no need''' anymore to run a special trimmer script and to import its output into <code>apertium-kaz-tat</code> manually.
To run tests, use <code>aq-regtest</code> utility from [[Apertium-quality]] tools. E.g. <pre>aq-regtest -d . kaz-tat http://wiki.apertium.org/wiki/Special:Export/Kazakh_and_Tatar/Postadvebs</pre>
 
   
  +
The same workflow applies for any other pair involving Kazakh or Tatar — if you need to change something in .lexc, .twol or .rlx files of these languages, you do so in apertium-kaz and apertium-tat directories respectively, compile monolingual packages, and then compile the translator.
== Done ==
 
   
  +
==== Adding language-pair-specific stems to the lexc files ====
; But keep an eye on this
 
  +
Sometimes we have to translate a word into Kazakh or Tatar with two or more words, e.g.:
 
  +
<pre><e><p><l>everywhere<s n="adv"/></l><r>барлық<b/>жерде<s n="adv"/></r></p></e></pre>
* Numerals
 
  +
In order to make it work, we will need to add ''барлық жерде'' as a single adverb in <code>kaz.lexc</code>, like this:
** kaz <num><subst>(<px3>) in fractions<ref>Currently whether it is in fractions or not is not taken into account</ref> = tat <num><subst>(<px3>)
 
  +
<pre>барлық% жерде:барлық% жерде ADV ; ! ""</pre>
** kaz <num><coll><advl> = tat <num><coll>
 
  +
But, certainly, these are two words in Kazakh — a determiner and a noun — and we don't want that the standalone Kazakh morphological analyzer would contain them as one word. In order to know that this entry was added for a specific language pair, we mark it with <code>Use/MT</code> at the end of the line:
** kaz <num><coll><subst> = tat <num><subst>
 
  +
<pre>барлық% жерде:барлық% жерде ADV ; ! "" Use/MT</pre>
 
  +
That way, we can collaborate on one single file across many pairs, but in the same time we are able to "clean up" the lexicon for standalone use if needed.
== Notes ==
 
 
<references/>
 
 
== See also ==
 
 
* [[/Pending tests]]
 
* [[/Regression tests]]
 
   
 
[[Category:Kazakh and Tatar|*]]
 
[[Category:Kazakh and Tatar|*]]

Latest revision as of 01:53, 10 March 2018

This is a language pair translating between Kazakh and Tatar. The pair is currently located in trunk.

General information[edit]

Demonstration[edit]

  • $ echo "Бұл аударушымен қазақша жазылған мәтіндерді татаршаға аударып оқуға болады." | apertium -d . kaz-tat
Бу тәрҗемәче белән казакъча язылган текстларны татарчага тәрҗемә итеп укуга була.

Installation[edit]

You will need:

If you are using a Debian-based distro, the easiest way to get those dependencies is to install them with apt-get from User:Tino Didriksen's repository.

Developers[edit]

Information on what remains to be done for this pair can be found at the /TODO list.

Development workflow[edit]

We work on the transducers (apertium-kaz and apertium-tat) individually, and use a special process to import to the pair transducers that contain only the words found in the bidix. The following documents this process.

Adding words[edit]

In order to add a new word (and its translation equivalent) to the Kazakh-Tatar translator, you have to do the following:

  1. add an entry in the bilingual dictionary — apertium-kaz-tat.kaz-tat.dix file in apertium-kaz-tat directory,
  2. add an entry in the Kazakh monolingual dictionary — apertium-kaz.kaz.lexc file, which, as the name indicates, is in the apertium-kaz directory,
  3. run make in apertium-kaz
  4. add an entry in the Tatar monolingual dictionary — apertium-tat.tat.lexc file in apertium-tat,
  5. run make in apertium-tat
  6. cd to apertium-kaz-tat and run make.

You have to have configured Kazakh-Tatar translator with the --with-lang1 and --with-lang2 options for the last step to work (see here for more details on this). It will fetch changed files automatically, trim them and compile them.

There is no need anymore to run a special trimmer script and to import its output into apertium-kaz-tat manually.

The same workflow applies for any other pair involving Kazakh or Tatar — if you need to change something in .lexc, .twol or .rlx files of these languages, you do so in apertium-kaz and apertium-tat directories respectively, compile monolingual packages, and then compile the translator.

Adding language-pair-specific stems to the lexc files[edit]

Sometimes we have to translate a word into Kazakh or Tatar with two or more words, e.g.:

<e><p><l>everywhere<s n="adv"/></l><r>барлық<b/>жерде<s n="adv"/></r></p></e>

In order to make it work, we will need to add барлық жерде as a single adverb in kaz.lexc, like this:

барлық% жерде:барлық% жерде ADV ; ! ""

But, certainly, these are two words in Kazakh — a determiner and a noun — and we don't want that the standalone Kazakh morphological analyzer would contain them as one word. In order to know that this entry was added for a specific language pair, we mark it with Use/MT at the end of the line:

барлық% жерде:барлық% жерде ADV ; ! "" Use/MT

That way, we can collaborate on one single file across many pairs, but in the same time we are able to "clean up" the lexicon for standalone use if needed.