Mentioned morpheme IDs in the documentation.

docs/_build/html/_sources/paradigms.rst.txt

@@ -188,6 +188,11 @@ Sometimes it is convenient to put certain stem characters into the paradigm. For
    gramm: gen,sg
    gloss: GEN.SG
 
+Morpheme IDs
+^^^^^^^^^^^^
+
+You can add an ``id`` field to morphemes and/or lexemes. IDs do not need to be unique and do not need to be assigned to each and every item. An analyzed word form will contain an ``id`` attribute if any of its parts had an ID. The value will contain the IDs of all its parts separated by a comma. Duplicate IDs will be truncated.
+
 Incorporated words
 ^^^^^^^^^^^^^^^^^^
 

docs/_build/html/paradigms.html

@@ -212,6 +212,10 @@ <h3>Stem parts<a class="headerlink" href="#stem-parts" title="Permalink to this
 </pre></div>
 </div>
 </div>
+<div class="section" id="morpheme-ids">
+<h3>Morpheme IDs<a class="headerlink" href="#morpheme-ids" title="Permalink to this headline">¶</a></h3>
+<p>You can add an <code class="docutils literal notranslate"><span class="pre">id</span></code> field to morphemes and/or lexemes. IDs do not need to be unique and do not need to be assigned to each and every item. An analyzed word form will contain an <code class="docutils literal notranslate"><span class="pre">id</span></code> attribute if any of its parts had an ID. The value will contain the IDs of all its parts separated by a comma. Duplicate IDs will be truncated.</p>
+</div>
 <div class="section" id="incorporated-words">
 <h3>Incorporated words<a class="headerlink" href="#incorporated-words" title="Permalink to this headline">¶</a></h3>
 <p>There are no tools for handling productive incorporation yet in <code class="docutils literal notranslate"><span class="pre">uniparser-morph</span></code>. Nevertheless, some incorporation can be accounted for in the paradigms. That can work if you have a limited number of words, e.g. pronominal clitics, that can be incorporated or orthographically fused with other words (hosts). Such words can be described as morphemes with a special <code class="docutils literal notranslate"><span class="pre">LEX</span></code> tag. Units with a <code class="docutils literal notranslate"><span class="pre">LEX</span></code> tag are processed as ordinary morphemes during parsing, but a separate “subword” analysis is added for each of them as one of the postprocessing steps. A <code class="docutils literal notranslate"><span class="pre">LEX</span></code> tag should look like <code class="docutils literal notranslate"><span class="pre">LEX:xxx:yyy</span></code>, where <code class="docutils literal notranslate"><span class="pre">xxx</span></code> is the lemma and <code class="docutils literal notranslate"><span class="pre">yyy</span></code> contains grammatical tags separated by a semicolon. (A semicolon is used so that a morpheme can have both <code class="docutils literal notranslate"><span class="pre">LEX</span></code> tags and regular tags, which are separated by a comma.)</p>

docs/paradigms.rst

@@ -188,6 +188,11 @@ Sometimes it is convenient to put certain stem characters into the paradigm. For
    gramm: gen,sg
    gloss: GEN.SG
 
+Morpheme IDs
+^^^^^^^^^^^^
+
+You can add an ``id`` field to morphemes and/or lexemes. IDs do not need to be unique and do not need to be assigned to each and every item. An analyzed word form will contain an ``id`` attribute if any of its parts had an ID. The value will contain the IDs of all its parts separated by a comma. Duplicate IDs will be truncated.
+
 Incorporated words
 ^^^^^^^^^^^^^^^^^^
 

setup.cfg

@@ -1,6 +1,6 @@
 [metadata]
 name = uniparser-morph
-version = 2.3.0
+version = 2.4.0
 author = Timofey Arkhangelskiy
 author_email = timarkh@gmail.com
 description = Rule-based, linguist-friendly (and rather slow) morphological analysis