diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
index 16f2979bedd2b9c0e204bdcb5df4881b0bf1c502..5627a1d717e78385d9270eb481bfedfd0c514e42 100644
--- a/.gitlab-ci.yml
+++ b/.gitlab-ci.yml
@@ -1,7 +1,17 @@
stages:
+ - test
- check_style
- build
+test_python:
+ stage: test
+ image: docker:18.09.7
+ services:
+ - docker:18.09.7-dind
+ script:
+ - docker build . -t wccl_test -f src/tests/docker/Dockerfile
+ - docker run --rm -v /tmp/test-out:/test/testdata/out wccl_test
+
build_image:
stage: build
image: docker:18.09.7
diff --git a/requirements-test.txt b/requirements-test.txt
new file mode 100644
index 0000000000000000000000000000000000000000..41ae417eb5a4236320cfb1942e821669479b2537
--- /dev/null
+++ b/requirements-test.txt
@@ -0,0 +1,6 @@
+corpus-ccl
+cclutils
+pudb
+pytest
+pytest-pudb
+wccl-annotator
\ No newline at end of file
diff --git a/src/libmwereader/mwereader.cpp b/src/libmwereader/mwereader.cpp
index a0acbcb66a930d3bfe7393b7fc6331126538123d..013841d1e78ca8dd9f70702f65a30c9dc3df5fd4 100644
--- a/src/libmwereader/mwereader.cpp
+++ b/src/libmwereader/mwereader.cpp
@@ -155,7 +155,7 @@ bool MWEReader::registered = TokenReader::register_path_reader<MWEReader>(
// move context position to next token after MWE elements
int curr_position = sentence_ctx->get_position();
if (curr_position + all.size() < sentence->size()) {
- sentence_ctx->set_position(sentence_ctx->get_position() + all.size());
+ sentence_ctx->set_position(sentence_ctx->get_position() + all.size() - 1);
}
}
@@ -171,6 +171,8 @@ bool MWEReader::registered = TokenReader::register_path_reader<MWEReader>(
// TODO: pass annotated sentence to methods
// AnnotatedSentencePtr ann_sentence = AnnotatedSentence::wrap_sentence(sentence);
int annotation_number = 0;
+ // keeps position of last annotated token
+ int last_ann_pos = -1;
SentenceContextPtr sc = boost::make_shared<Wccl::SentenceContext>(sentence);
for (sc->goto_start(); sc->is_current_inside(); sc->advance())
@@ -184,6 +186,10 @@ bool MWEReader::registered = TokenReader::register_path_reader<MWEReader>(
if(lex.is_disamb()){
std::string base = lex.lemma_utf8();
const MWEIndex::luvec& potential = mwe_index_.get_potential_lu(base);
+ // variables holding values for selected lu
+ LexicalUnit::Ptr sel_pLU;
+ std::set<int> sel_positions;
+ int sel_head;
BOOST_FOREACH (LexicalUnit::Ptr pLU, potential)
{
bool ok = true;
@@ -202,20 +208,45 @@ bool MWEReader::registered = TokenReader::register_path_reader<MWEReader>(
int head;
bool is_here = pLU->IsHere(sc, positions, head);
if(is_here) {
- if (annotate) {
- add_mwe_channel(
- sc, head, positions,
- ++annotation_number,
- pLU->get_base());
- }
- else {
- sc = clone_sentence_add_mwe(
- sc, head, positions,
- pLU->get_base());
+ if (positions.size() > sel_positions.size()) {
+ sel_pLU = pLU;
+ sel_positions = positions;
+ sel_head = head;
}
}
}
}
+ if (sel_pLU) {
+ int first_curr_pos;
+ int last_curr_pos;
+ if(!sel_positions.empty())
+ first_curr_pos = *sel_positions.begin();
+ // annotate only if tokens have not been already
+ // anotated; Below check relies on first token and
+ // prevents from re-anotating the last token in the
+ // sentence (grom previous annotating)
+ if (first_curr_pos > last_ann_pos) {
+ if (annotate) {
+ add_mwe_channel(
+ sc,
+ sel_head,
+ sel_positions,
+ ++annotation_number,
+ sel_pLU->get_base()
+ );
+ }
+ else {
+ sc = clone_sentence_add_mwe(
+ sc,
+ sel_head,
+ sel_positions,
+ sel_pLU->get_base()
+ );
+ }
+ last_curr_pos = *sel_positions.rbegin();
+ last_ann_pos = last_curr_pos;
+ }
+ }
}
}
}
diff --git a/src/tests/docker/Dockerfile b/src/tests/docker/Dockerfile
new file mode 100644
index 0000000000000000000000000000000000000000..efb5c582ccc096194f4ff155132d16880948af45
--- /dev/null
+++ b/src/tests/docker/Dockerfile
@@ -0,0 +1,50 @@
+FROM clarinpl/python:3.6
+
+RUN apt-get update && apt-get install -y \
+ libxml++2.6-dev \
+ libloki-dev \
+ libboost-all-dev \
+ libicu-dev \
+ libffi-dev \
+ libssl-dev \
+ libxml2-utils \
+ cmake \
+ swig \
+ pwrutils \
+ gdebi-core \
+ antlr \
+ libantlr-dev \
+ default-jdk \
+ git
+
+RUN mkdir -p /home/install
+WORKDIR /home/install
+
+RUN bash -c "wget -q -O - http://apt.clarin-pl.eu/KEY.gpg | apt-key add -"
+RUN bash -c "echo 'deb https://apt.clarin-pl.eu/ /' > /etc/apt/sources.list.d/clarin.list"
+
+# Install corpus2 and corpus2mwe
+RUN apt-get update && apt-get install -y \
+ corpus2-python3.6 \
+ corpus2mwe-python3.6
+
+# copy src
+COPY ./ ./wccl
+
+# Install pip test dependencies
+RUN pip install -r wccl/requirements-test.txt
+
+# copy tests
+COPY ./src/tests/python-tests /test
+
+# Install wccl from this local repo
+WORKDIR /home/install
+RUN mkdir wccl/src/build && \
+ cd wccl/src/build && \
+ cmake .. && \
+ make -j 8 && \
+ make install && \
+ ldconfig
+
+WORKDIR /test
+CMD ["python", "-m", "pytest"]
diff --git a/src/tests/python-tests/__init__.py b/src/tests/python-tests/__init__.py
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391
diff --git a/src/tests/python-tests/ccl_ann_test.py b/src/tests/python-tests/ccl_ann_test.py
new file mode 100644
index 0000000000000000000000000000000000000000..60fb862702e92d4b70b71fe49841f8727eae668f
--- /dev/null
+++ b/src/tests/python-tests/ccl_ann_test.py
@@ -0,0 +1,177 @@
+from typing import Any, List, Tuple
+
+from corpus_ccl import token_utils as tokutils
+
+
+class CclIndex(object):
+ """
+ Class builds indices useful for convenient referencing tokens from CCL
+ document.
+ """
+
+ def __init__(self, ccl_doc: "corpus2.DocumentPtr"):
+ self._doc = ccl_doc
+ self._tok_idx, self._sent_idx = self._make_indices()
+
+ def get_token(self, par_idx: int, sent_idx: int, tok_idx: int) -> "corpus2.Token":
+ """
+ Returns specified token. Indices are counts of document entities.
+ Counting starts from 0 (e.g. get_token(0, 1, 2) will return third token from
+ second sentence, located in first paragraph).
+ """
+ return self._tok_idx[par_idx][sent_idx][tok_idx]
+
+ def get_sent_by_token_idx(
+ self, token_idx: Tuple[int, int, int]
+ ) -> "corpus2.Sentence":
+ return self._sent_idx[token_idx]
+
+ def _make_indices(self) -> Tuple[dict, dict]:
+ """
+ Generates nested dict of tokens from CCL document, with structure
+ corresponding to such document. Tokens can be retrieved from generated
+ dict using indices of paragraph, sentence and token.
+ Note: indexing starts from 0 (e.g. d[0][1][2] refers to third token from
+ second sentence, located in first paragraph).
+
+ Args:
+ doc: CCL document
+
+ Returns:
+ (dict, dict), where first is a nested dict containing tokens
+ and the second is an index of sentences referred by the tokens.
+ """
+ d = {}
+ sents = {}
+ for pidx, p in enumerate(self._doc.paragraphs()):
+ pd = {}
+ for sidx, s in enumerate(p.sentences()):
+ sd = {}
+ for tidx, t in enumerate(s.tokens()):
+ sd[tidx] = t
+ sents[(pidx, sidx, tidx)] = s
+ pd[sidx] = sd
+ d[pidx] = pd
+ return d, sents
+
+
+def assert_has_ann(
+ tokens_idx_list: List[Tuple[int, int, int]],
+ doc_idx: CclIndex,
+ ann_names: Tuple[str, str],
+ base_ann_value,
+) -> bool:
+ """
+ Main assertion method to check correctness of annotation to one or more
+ adjecent tokens.
+
+ Checks whether one or more tokens have been annotated with certain
+ annotation. It verifies:
+ 1) whether base annotation value is specified for exactly one token (head)
+ and there is no other bases for given annotation channel,
+ 2) in case of more than one tokens: if they belongs to the same annotation
+ (if have same channel value, index)
+ """
+ ann_chan_name, ann_chan_base_name = ann_names
+ has_base = False
+ found_bases = []
+ for token_idx in tokens_idx_list:
+ found_base = get_prop_val(
+ doc_idx.get_token(*token_idx),
+ ann_chan_base_name,
+ )
+ if found_base is not None:
+ found_bases.append(found_base)
+ if not found_bases:
+ toks_str = [doc_idx.get_token(*ti).orth_utf8() for ti in tokens_idx_list]
+ raise AssertionError(
+ f"None of specified tokens with ids={tokens_idx_list} ({toks_str}) "
+ f"contains requested base name '{base_ann_value}' "
+ f"for annotation channel '{ann_chan_base_name}'"
+ )
+ if len(found_bases) > 1:
+ toks_str = [doc_idx.get_token(*ti).orth_utf8() for ti in tokens_idx_list]
+ raise AssertionError(
+ f"Found more than one base value in specified tokens with ids="
+ f"{tokens_idx_list} ({toks_str}) for annotation channel '{ann_chan_base_name}':"
+ f"{found_bases}"
+ )
+ if found_bases[0] != base_ann_value:
+ toks_str = [doc_idx.get_token(*ti).orth_utf8() for ti in tokens_idx_list]
+ raise AssertionError(
+ f"Found incorrect base value in specified tokens with ids="
+ f"{tokens_idx_list} ({toks_str}) for annotation channel '{ann_chan_base_name}':"
+ f"'{found_bases[0]}' (expected: {base_ann_value})"
+ )
+
+ ann_chan_idx = 0
+ for token_idx in tokens_idx_list:
+ tok = doc_idx.get_token(*token_idx)
+ curr_ann_chan_idx = get_chan_ann_id(
+ tok,
+ doc_idx.get_sent_by_token_idx(token_idx),
+ ann_chan_name,
+ )
+
+ if not curr_ann_chan_idx:
+ raise AssertionError(
+ f"token with id={token_idx} ('{tok.orth_utf8()}') does not have "
+ f"set annotation for annotation channel '{ann_chan_name}'"
+ )
+ if not ann_chan_idx:
+ ann_chan_idx = curr_ann_chan_idx
+ else:
+ if ann_chan_idx != curr_ann_chan_idx:
+ raise AssertionError(
+ "Given tokens belongs to different annotation channels for "
+ f"annotation '{ann_chan_name}' (found channels: "
+ f"{ann_chan_idx} , {curr_ann_chan_idx})"
+ )
+
+
+def get_prop_val(
+ tok: "corpus2.Token",
+ key_name: str,
+) -> Any:
+ """
+ Returns value of property from given token.
+ """
+ if tok.has_metadata():
+ md = tok.get_metadata()
+ else:
+ return None
+ keys = md.attributes().keys()
+ val = md.get_attribute(key_name)
+ return val
+
+
+def get_chan_ann_id(token: "corpus2.Token", sent: "corpus2.Sentence", ann) -> int:
+ """
+ Returns value of channel annotation in sentence for given token.
+ If 0 is returned, then it mean that there is no annotation for given token.
+ """
+ return tokutils.get_annotation(sent, token, ann, 0)
+
+
+def expand_tok_idx(tokens_idx_def):
+ """
+ Small function to generate list of identifiers based on shortened
+ description of tokens range. E.g. converts (0, 0, (1, 3)) into
+ [
+ (0, 0, 1),
+ (0, 0, 2),
+ (0, 0, 3)
+ ].
+ Notes:
+ - it includes both ends of a range!
+ - if third element represents a range of tokens, it must be a tuple.
+ """
+ if isinstance(tokens_idx_def, tuple) and isinstance(tokens_idx_def[2], tuple):
+ par_idx, sent_idx, tok_def = tokens_idx_def
+ exp = []
+ tok_first, tok_last = tok_def
+ for i in range(tok_first, tok_last + 1):
+ exp.append((par_idx, sent_idx, i))
+ return exp
+ else:
+ return tokens_idx_def
diff --git a/src/tests/python-tests/test_wccl.py b/src/tests/python-tests/test_wccl.py
new file mode 100644
index 0000000000000000000000000000000000000000..4e88cfd8cfe91ceee027124177faa60014fe29b0
--- /dev/null
+++ b/src/tests/python-tests/test_wccl.py
@@ -0,0 +1,197 @@
+from collections import namedtuple
+import sys
+import os
+from typing import List
+
+import pytest
+
+import cclutils as ccl
+from wccl_annotator.wccl_annotator import WcclAnnotator
+
+import ccl_ann_test
+
+
+TEST_ROOT_DIR = "/test/testdata"
+TEST_OUT_DIR = "/test/testdata/out"
+TEST_ANN_CHAN = "test_ann"
+TEST_ANN_CHAN_BASE = "test_ann_base"
+TEST_ANNS = (TEST_ANN_CHAN, TEST_ANN_CHAN_BASE)
+
+if not os.path.isdir(TEST_OUT_DIR):
+ os.mkdir(TEST_OUT_DIR)
+
+# description of annotation to detect; contains information about location
+# in document, channel and expected base value
+AnnCfg = namedtuple("AnnCfg", "tokens ann_names base_ann_value")
+
+
+@pytest.mark.parametrize(
+ "ccl_file, wccl_dict, out_file, tagset, expected_anns",
+ [
+ pytest.param(
+ f"{TEST_ROOT_DIR}/doc_01.xml",
+ f"{TEST_ROOT_DIR}/wccl_dict_01.xml",
+ f"{TEST_OUT_DIR}/doc_01_annotated.xml",
+ "nkjp",
+ [
+ AnnCfg((0, 0, (0, 1)), TEST_ANNS, "plac zabaw"),
+ AnnCfg([(0, 0, 3)], TEST_ANNS, "tenis"),
+ AnnCfg((0, 0, (6, 7)), TEST_ANNS, "all inclusive"),
+ ],
+ id="basic-test-word-and-phrases",
+ ),
+ pytest.param(
+ f"{TEST_ROOT_DIR}/doc_02.xml",
+ f"{TEST_ROOT_DIR}/wccl_dict_02.xml",
+ f"{TEST_OUT_DIR}/doc_02_annotated.xml",
+ "nkjp",
+ [
+ AnnCfg([(0, 0, 0)], TEST_ANNS, "ZO"),
+ AnnCfg((0, 0, (1, 2)), TEST_ANNS, "all inclusive"),
+ AnnCfg((0, 0, (4, 5)), TEST_ANNS, "plac zabaw"),
+ ],
+ id="adjacent-anns",
+ ),
+ pytest.param(
+ f"{TEST_ROOT_DIR}/doc_03.xml",
+ f"{TEST_ROOT_DIR}/wccl_dict_03a.xml",
+ f"{TEST_OUT_DIR}/doc_03a_annotated.xml",
+ "nkjp",
+ [
+ AnnCfg((0, 0, (1, 2)), TEST_ANNS, "plac zabaw"),
+ ],
+ id="dict-terms-inclusion-longer-first",
+ ),
+ pytest.param(
+ f"{TEST_ROOT_DIR}/doc_03.xml",
+ f"{TEST_ROOT_DIR}/wccl_dict_03b.xml",
+ f"{TEST_OUT_DIR}/doc_03b_annotated.xml",
+ "nkjp",
+ [
+ AnnCfg((0, 0, (1, 2)), TEST_ANNS, "plac zabaw"),
+ ],
+ id="dict-terms-inclusion-shorter-first",
+ ),
+ pytest.param(
+ f"{TEST_ROOT_DIR}/doc_03b.xml",
+ f"{TEST_ROOT_DIR}/wccl_dict_03a.xml",
+ f"{TEST_OUT_DIR}/doc_03a_annotated_wb.xml",
+ "nkjp",
+ [
+ AnnCfg((0, 0, (1, 2)), TEST_ANNS, "plac zabaw"),
+ ],
+ id="dict-terms-inclusion-longer-first-and-adjecent-word-before",
+ ),
+ pytest.param(
+ f"{TEST_ROOT_DIR}/doc_03b.xml",
+ f"{TEST_ROOT_DIR}/wccl_dict_03b.xml",
+ f"{TEST_OUT_DIR}/doc_03b_annotated_wb.xml",
+ "nkjp",
+ [
+ AnnCfg([(0, 0, 0)], TEST_ANNS, "plac"),
+ AnnCfg((0, 0, (1, 2)), TEST_ANNS, "plac zabaw"),
+ ],
+ id="dict-terms-inclusion-shorter-first-and-adjecent-word-before",
+ ),
+ pytest.param(
+ f"{TEST_ROOT_DIR}/doc_03c.xml",
+ f"{TEST_ROOT_DIR}/wccl_dict_03a.xml",
+ f"{TEST_OUT_DIR}/doc_03c_annotated_wa.xml",
+ "nkjp",
+ [
+ AnnCfg((0, 0, (1, 2)), TEST_ANNS, "plac zabaw"),
+ ],
+ id="dict-terms-inclusion-longer-first-and-adjecent-word-after",
+ ),
+ pytest.param(
+ f"{TEST_ROOT_DIR}/doc_03c.xml",
+ f"{TEST_ROOT_DIR}/wccl_dict_03b.xml",
+ f"{TEST_OUT_DIR}/doc_03c_annotated_wa.xml",
+ "nkjp",
+ [
+ AnnCfg((0, 0, (1, 2)), TEST_ANNS, "plac zabaw"),
+ ],
+ id="dict-terms-inclusion-shorter-first-and-adjecent-word-after",
+ ),
+ pytest.param(
+ f"{TEST_ROOT_DIR}/doc_03b.xml",
+ f"{TEST_ROOT_DIR}/wccl_dict_04a.xml",
+ f"{TEST_OUT_DIR}/doc_03b-04a_annotated.xml",
+ "nkjp",
+ [
+ AnnCfg([(0, 0, 0)], TEST_ANNS, "plac"),
+ AnnCfg((0, 0, (1, 2)), TEST_ANNS, "plac zabaw"),
+ ],
+ id="dict-terms-double-inclusion-longer-first",
+ ),
+ pytest.param(
+ f"{TEST_ROOT_DIR}/doc_03b.xml",
+ f"{TEST_ROOT_DIR}/wccl_dict_04b.xml",
+ f"{TEST_OUT_DIR}/doc_03b-04b_annotated.xml",
+ "nkjp",
+ [
+ AnnCfg([(0, 0, 0)], TEST_ANNS, "plac"),
+ AnnCfg((0, 0, (1, 2)), TEST_ANNS, "plac zabaw"),
+ ],
+ id="dict-terms-double-inclusion-shorter-first",
+ ),
+ pytest.param(
+ f"{TEST_ROOT_DIR}/doc_03e.xml",
+ f"{TEST_ROOT_DIR}/wccl_dict_04b.xml",
+ f"{TEST_OUT_DIR}/doc_03e-04b_annotated.xml",
+ "nkjp",
+ [
+ AnnCfg([(0, 0, 0)], TEST_ANNS, "plac"),
+ AnnCfg((0, 0, (1, 2)), TEST_ANNS, "plac zabaw"),
+ ],
+ id="dict-terms-double-inclusion-shorter-first-both-words",
+ ),
+ pytest.param(
+ f"{TEST_ROOT_DIR}/doc_03d.xml",
+ f"{TEST_ROOT_DIR}/wccl_dict_04b.xml",
+ f"{TEST_OUT_DIR}/doc_03b-04b_annotated-alt.xml",
+ "nkjp",
+ [
+ AnnCfg([(0, 0, 0)], TEST_ANNS, "plac"),
+ AnnCfg((0, 0, (1, 2)), TEST_ANNS, "plac zabaw"),
+ ],
+ id="dict-terms-double-inclusion-shorter-first-additional-last-token",
+ ),
+ pytest.param(
+ f"{TEST_ROOT_DIR}/doc_03b.xml",
+ f"{TEST_ROOT_DIR}/wccl_dict_04c.xml",
+ f"{TEST_OUT_DIR}/doc_03b-04c_annotated.xml",
+ "nkjp",
+ [
+ AnnCfg([(0, 0, 0)], TEST_ANNS, "plac"),
+ AnnCfg((0, 0, (1, 2)), TEST_ANNS, "plac zabaw"),
+ ],
+ id="dict-terms-double-inclusion-shifted-head",
+ ),
+ ],
+)
+def test_wccl_annotations(
+ ccl_file: str,
+ wccl_dict: str,
+ out_file: str,
+ tagset: str,
+ expected_anns: List[AnnCfg],
+):
+ WcclAnnotator().process(
+ ccl_file,
+ out_file,
+ annotation=TEST_ANN_CHAN,
+ wccl_dict=wccl_dict,
+ tagset=tagset,
+ )
+ # WcclAnnotator stores annotated document, so we read it from file
+ doc = ccl.read(out_file)
+ doc_idx = ccl_ann_test.CclIndex(doc)
+ for exp in expected_anns:
+ tokens_indices = ccl_ann_test.expand_tok_idx(exp.tokens)
+ ccl_ann_test.assert_has_ann(
+ tokens_indices,
+ doc_idx,
+ exp.ann_names,
+ exp.base_ann_value,
+ )
diff --git a/src/tests/python-tests/testdata/doc_01.xml b/src/tests/python-tests/testdata/doc_01.xml
new file mode 100644
index 0000000000000000000000000000000000000000..26917ba00a2d626434dde82bbce5f1e30add5252
--- /dev/null
+++ b/src/tests/python-tests/testdata/doc_01.xml
@@ -0,0 +1,40 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE chunkList SYSTEM "ccl.dtd">
+<chunkList>
+<chunk id="1" type="p">
+<sentence id="1">
+<tok>
+<orth>plac</orth>
+<lex disamb="1"><base>plac</base><ctag>subst:sg:nom:m3</ctag></lex>
+</tok>
+<tok>
+<orth>zabaw</orth>
+<lex disamb="1"><base>zabawa</base><ctag>subst:pl:gen:f</ctag></lex>
+</tok>
+<tok>
+<orth>i</orth>
+<lex disamb="1"><base>i</base><ctag>conj</ctag></lex>
+</tok>
+<tok>
+<orth>tenis</orth>
+<lex disamb="1"><base>tenis</base><ctag>subst:sg:nom:m3</ctag></lex>
+</tok>
+<tok>
+<orth>stołowy</orth>
+<lex disamb="1"><base>stołowy</base><ctag>adj:sg:nom:m3:pos</ctag></lex>
+</tok>
+<tok>
+<orth>z</orth>
+<lex disamb="1"><base>z</base><ctag>prep:gen:nwok</ctag></lex>
+</tok>
+<tok>
+<orth>all</orth>
+<lex disamb="1"><base>all</base><ctag>subst:sg:nom:m3</ctag></lex>
+</tok>
+<tok>
+<orth>inclusive</orth>
+<lex disamb="1"><base>inclusive</base><ctag>adj:sg:nom:m3:pos</ctag></lex>
+</tok>
+</sentence>
+</chunk>
+</chunkList>
diff --git a/src/tests/python-tests/testdata/doc_01_annotated.xml b/src/tests/python-tests/testdata/doc_01_annotated.xml
new file mode 100644
index 0000000000000000000000000000000000000000..35827bb6a4c027f03da5e80406528460bae7b7c8
--- /dev/null
+++ b/src/tests/python-tests/testdata/doc_01_annotated.xml
@@ -0,0 +1,51 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE chunkList SYSTEM "ccl.dtd">
+<chunkList>
+ <chunk id="1">
+ <sentence id="1">
+ <tok>
+ <orth>plac</orth>
+ <lex disamb="1"><base>plac</base><ctag>subst:sg:nom:m3</ctag></lex>
+ <ann chan="test_ann">1</ann>
+ <prop key="test_ann_base">plac zbaw</prop>
+ </tok>
+ <tok>
+ <orth>zabaw</orth>
+ <lex disamb="1"><base>zabawa</base><ctag>subst:pl:gen:f</ctag></lex>
+ <ann chan="test_ann">1</ann>
+ </tok>
+ <tok>
+ <orth>i</orth>
+ <lex disamb="1"><base>i</base><ctag>conj</ctag></lex>
+ <ann chan="test_ann">0</ann>
+ </tok>
+ <tok>
+ <orth>tenis</orth>
+ <lex disamb="1"><base>tenis</base><ctag>subst:sg:nom:m3</ctag></lex>
+ <ann chan="test_ann">2</ann>
+ <prop key="test_ann_base">tenis stołowy</prop>
+ </tok>
+ <tok>
+ <orth>stołowy</orth>
+ <lex disamb="1"><base>stołowy</base><ctag>adj:sg:nom:m3:pos</ctag></lex>
+ <ann chan="test_ann">2</ann>
+ </tok>
+ <tok>
+ <orth>z</orth>
+ <lex disamb="1"><base>z</base><ctag>prep:gen:nwok</ctag></lex>
+ <ann chan="test_ann">0</ann>
+ </tok>
+ <tok>
+ <orth>all</orth>
+ <lex disamb="1"><base>all</base><ctag>subst:sg:nom:m3</ctag></lex>
+ <ann chan="test_ann">3</ann>
+ <prop key="test_ann_base">all inclusive</prop>
+ </tok>
+ <tok>
+ <orth>inclusive</orth>
+ <lex disamb="1"><base>inclusive</base><ctag>adj:sg:nom:m3:pos</ctag></lex>
+ <ann chan="test_ann">3</ann>
+ </tok>
+ </sentence>
+ </chunk>
+</chunkList>
diff --git a/src/tests/python-tests/testdata/doc_02.xml b/src/tests/python-tests/testdata/doc_02.xml
new file mode 100644
index 0000000000000000000000000000000000000000..ee01d6675a6763aa7de4d6b9df72d25ecf1a0158
--- /dev/null
+++ b/src/tests/python-tests/testdata/doc_02.xml
@@ -0,0 +1,32 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE chunkList SYSTEM "ccl.dtd">
+<chunkList>
+<chunk id="1" type="p">
+<sentence id="1">
+<tok>
+<orth>z</orth>
+<lex disamb="1"><base>z</base><ctag>prep:gen:nwok</ctag></lex>
+</tok>
+<tok>
+<orth>all</orth>
+<lex disamb="1"><base>all</base><ctag>subst:sg:nom:m3</ctag></lex>
+</tok>
+<tok>
+<orth>inclusive</orth>
+<lex disamb="1"><base>inclusive</base><ctag>adj:sg:nom:m3:pos</ctag></lex>
+</tok>
+<tok>
+<orth>i</orth>
+<lex disamb="1"><base>i</base><ctag>conj</ctag></lex>
+</tok>
+<tok>
+<orth>plac</orth>
+<lex disamb="1"><base>plac</base><ctag>subst:sg:nom:m3</ctag></lex>
+</tok>
+<tok>
+<orth>zabaw</orth>
+<lex disamb="1"><base>zabawa</base><ctag>subst:pl:gen:f</ctag></lex>
+</tok>
+</sentence>
+</chunk>
+</chunkList>
diff --git a/src/tests/python-tests/testdata/doc_03.xml b/src/tests/python-tests/testdata/doc_03.xml
new file mode 100644
index 0000000000000000000000000000000000000000..230b0f8b7cbcff0f6e060f413200a4a0c07346e5
--- /dev/null
+++ b/src/tests/python-tests/testdata/doc_03.xml
@@ -0,0 +1,20 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE chunkList SYSTEM "ccl.dtd">
+<chunkList>
+<chunk id="1" type="p">
+<sentence id="1">
+<tok>
+<orth>z</orth>
+<lex disamb="1"><base>z</base><ctag>prep:gen:nwok</ctag></lex>
+</tok>
+<tok>
+<orth>placem</orth>
+<lex disamb="1"><base>plac</base><ctag>subst:sg:nom:m3</ctag></lex>
+</tok>
+<tok>
+<orth>zabaw</orth>
+<lex disamb="1"><base>zabawa</base><ctag>subst:pl:gen:f</ctag></lex>
+</tok>
+</sentence>
+</chunk>
+</chunkList>
diff --git a/src/tests/python-tests/testdata/doc_03b.xml b/src/tests/python-tests/testdata/doc_03b.xml
new file mode 100644
index 0000000000000000000000000000000000000000..0cc5fdab2c8c15ead2b484c4b6295e679067d7d6
--- /dev/null
+++ b/src/tests/python-tests/testdata/doc_03b.xml
@@ -0,0 +1,20 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE chunkList SYSTEM "ccl.dtd">
+<chunkList>
+<chunk id="1" type="p">
+<sentence id="1">
+<tok>
+<orth>plac</orth>
+<lex disamb="1"><base>plac</base><ctag>subst:sg:nom:m3</ctag></lex>
+</tok>
+<tok>
+<orth>placem</orth>
+<lex disamb="1"><base>plac</base><ctag>subst:sg:nom:m3</ctag></lex>
+</tok>
+<tok>
+<orth>zabaw</orth>
+<lex disamb="1"><base>zabawa</base><ctag>subst:pl:gen:f</ctag></lex>
+</tok>
+</sentence>
+</chunk>
+</chunkList>
diff --git a/src/tests/python-tests/testdata/doc_03c.xml b/src/tests/python-tests/testdata/doc_03c.xml
new file mode 100644
index 0000000000000000000000000000000000000000..54e7e3c084708c809de9855a64e30771349f3c18
--- /dev/null
+++ b/src/tests/python-tests/testdata/doc_03c.xml
@@ -0,0 +1,24 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE chunkList SYSTEM "ccl.dtd">
+<chunkList>
+<chunk id="1" type="p">
+<sentence id="1">
+<tok>
+<orth>z</orth>
+<lex disamb="1"><base>z</base><ctag>prep:gen:nwok</ctag></lex>
+</tok>
+<tok>
+<orth>placem</orth>
+<lex disamb="1"><base>plac</base><ctag>subst:sg:nom:m3</ctag></lex>
+</tok>
+<tok>
+<orth>zabaw</orth>
+<lex disamb="1"><base>zabawa</base><ctag>subst:pl:gen:f</ctag></lex>
+</tok>
+<tok>
+<orth>plac</orth>
+<lex disamb="1"><base>plac</base><ctag>subst:sg:nom:m3</ctag></lex>
+</tok>
+</sentence>
+</chunk>
+</chunkList>
diff --git a/src/tests/python-tests/testdata/doc_03d.xml b/src/tests/python-tests/testdata/doc_03d.xml
new file mode 100644
index 0000000000000000000000000000000000000000..ac667e69428875cc3d2532700433d22455ddc6a4
--- /dev/null
+++ b/src/tests/python-tests/testdata/doc_03d.xml
@@ -0,0 +1,24 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE chunkList SYSTEM "ccl.dtd">
+<chunkList>
+<chunk id="1" type="p">
+<sentence id="1">
+<tok>
+<orth>plac</orth>
+<lex disamb="1"><base>plac</base><ctag>subst:sg:nom:m3</ctag></lex>
+</tok>
+<tok>
+<orth>placem</orth>
+<lex disamb="1"><base>plac</base><ctag>subst:sg:nom:m3</ctag></lex>
+</tok>
+<tok>
+<orth>zabaw</orth>
+<lex disamb="1"><base>zabawa</base><ctag>subst:pl:gen:f</ctag></lex>
+</tok>
+<tok>
+<orth>z</orth>
+<lex disamb="1"><base>z</base><ctag>prep:gen:nwok</ctag></lex>
+</tok>
+</sentence>
+</chunk>
+</chunkList>
diff --git a/src/tests/python-tests/testdata/doc_03e.xml b/src/tests/python-tests/testdata/doc_03e.xml
new file mode 100644
index 0000000000000000000000000000000000000000..02d3953b8211a6c29fa1c9fb18b111e5397f9c88
--- /dev/null
+++ b/src/tests/python-tests/testdata/doc_03e.xml
@@ -0,0 +1,24 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE chunkList SYSTEM "ccl.dtd">
+<chunkList>
+<chunk id="1" type="p">
+<sentence id="1">
+<tok>
+<orth>plac</orth>
+<lex disamb="1"><base>plac</base><ctag>subst:sg:nom:m3</ctag></lex>
+</tok>
+<tok>
+<orth>placem</orth>
+<lex disamb="1"><base>plac</base><ctag>subst:sg:nom:m3</ctag></lex>
+</tok>
+<tok>
+<orth>zabaw</orth>
+<lex disamb="1"><base>zabawa</base><ctag>subst:pl:gen:f</ctag></lex>
+</tok>
+<tok>
+<orth>zabawa</orth>
+<lex disamb="1"><base>zabawa</base><ctag>subst:pl:gen:f</ctag></lex>
+</tok>
+</sentence>
+</chunk>
+</chunkList>
diff --git a/src/tests/python-tests/testdata/wccl_dict_01.xml b/src/tests/python-tests/testdata/wccl_dict_01.xml
new file mode 100644
index 0000000000000000000000000000000000000000..da59cb7bd8e5956fe36fa19153cb8ddc9e5fda95
--- /dev/null
+++ b/src/tests/python-tests/testdata/wccl_dict_01.xml
@@ -0,0 +1,103 @@
+<units_description tagset="nkjp">
+
+ <mwegroup class="neo" id="SubstSubst" name="SubstSubst" type="fix">
+
+ <condition>
+ and(
+ equal(base[0],$s:A),
+ equal(base[1],$s:B),
+ inter(base[0],$s:A), inter(class[0],{subst}),
+ inter(base[1],$s:B), inter(class[1],{subst}),
+ setvar($Pos0, 0),
+ setvar($Pos1, 1),
+ setvar($Head, 0)
+ )
+ </condition>
+
+
+ <instances>
+
+ <MWE base="plac zabaw">
+ <var name="A">plac</var>
+<var name="B">zabawa</var>
+ </MWE>
+
+ </instances>
+
+ </mwegroup>
+
+ <mwegroup class="neo" id="Subst" name="Subst" type="fix">
+
+ <condition>
+ and(
+ equal(base[0],$s:A),
+ inter(base[0],$s:A), inter(class[0],{subst}),
+ setvar($Pos0, 0),
+ setvar($Head, 0)
+ )
+ </condition>
+
+
+ <instances>
+
+ <MWE base="tenis">
+ <var name="A">tenis</var>
+ </MWE>
+
+ </instances>
+
+ </mwegroup>
+
+ <mwegroup class="neo" id="SubstIgn" name="SubstIgn" type="fix">
+
+ <condition>
+ and(
+ equal(base[0],$s:A),
+ equal(base[1],$s:B),
+ inter(base[0],$s:A), inter(class[0],{subst}),
+ inter(base[1],$s:B), inter(class[1],{adj}),
+ setvar($Pos0, 0),
+ setvar($Pos1, 1),
+ setvar($Head, 0)
+ )
+ </condition>
+
+
+ <instances>
+ <MWE base="all inclusive">
+ <var name="A">all</var>
+<var name="B">inclusive</var>
+ </MWE>
+
+ </instances>
+
+ </mwegroup>
+
+ <mwegroup class="neo" id="SubstInterpIgn" name="SubstInterpIgn" type="fix">
+
+ <condition>
+ and(
+ equal(base[0],$s:A),
+ equal(base[1],$s:B),
+ equal(base[2],$s:C),
+ inter(base[0],$s:A), inter(class[0],{subst}),
+ inter(base[1],$s:B), inter(class[1],{interp}),
+ inter(base[2],$s:C), inter(class[2],{ign}),
+ setvar($Pos0, 0),
+ setvar($Pos1, 1),
+ setvar($Pos2, 2),
+ setvar($Head, 0)
+ )
+ </condition>
+
+
+ <instances>
+ <MWE base="all - inclusive">
+ <var name="A">all</var>
+<var name="B">-</var>
+<var name="C">inclusive</var>
+ </MWE>
+ </instances>
+</mwegroup>
+
+ </units_description>
diff --git a/src/tests/python-tests/testdata/wccl_dict_02.xml b/src/tests/python-tests/testdata/wccl_dict_02.xml
new file mode 100644
index 0000000000000000000000000000000000000000..e1623d7cc263fb605a991d2c325e077a1704aa13
--- /dev/null
+++ b/src/tests/python-tests/testdata/wccl_dict_02.xml
@@ -0,0 +1,127 @@
+<units_description tagset="nkjp">
+
+ <mwegroup class="neo" id="SubstAdj" name="SubstAdj" type="fix">
+
+ <condition>
+ and(
+ equal(base[0],$s:A),
+ equal(base[1],$s:B),
+ inter(base[0],$s:A), inter(class[0],{subst}),
+ inter(base[1],$s:B), inter(class[1],{adj}),
+ setvar($Pos0, 0),
+ setvar($Pos1, 1),
+ setvar($Head, 0)
+ )
+ </condition>
+
+
+ <instances>
+
+ <MWE base="tenis stołowy">
+ <var name="A">tenis</var>
+<var name="B">stołowy</var>
+ </MWE>
+
+ </instances>
+
+ </mwegroup>
+
+ <mwegroup class="neo" id="SubstSubst" name="SubstSubst" type="fix">
+
+ <condition>
+ and(
+ equal(base[0],$s:A),
+ equal(base[1],$s:B),
+ inter(base[0],$s:A), inter(class[0],{subst}),
+ inter(base[1],$s:B), inter(class[1],{subst}),
+ setvar($Pos0, 0),
+ setvar($Pos1, 1),
+ setvar($Head, 0)
+ )
+ </condition>
+
+
+ <instances>
+
+ <MWE base="plac zabaw">
+ <var name="A">plac</var>
+<var name="B">zabawa</var>
+ </MWE>
+
+ </instances>
+
+ </mwegroup>
+
+ <mwegroup class="neo" id="SubstIgn" name="SubstIgn" type="fix">
+
+ <condition>
+ and(
+ equal(base[0],$s:A),
+ equal(base[1],$s:B),
+ inter(base[0],$s:A), inter(class[0],{subst}),
+ inter(base[1],$s:B), inter(class[1],{adj}),
+ setvar($Pos0, 0),
+ setvar($Pos1, 1),
+ setvar($Head, 0)
+ )
+ </condition>
+
+
+ <instances>
+ <MWE base="all inclusive">
+ <var name="A">all</var>
+<var name="B">inclusive</var>
+ </MWE>
+
+ </instances>
+
+ </mwegroup>
+
+ <mwegroup class="neo" id="SubstInterpIgn" name="SubstInterpIgn" type="fix">
+
+ <condition>
+ and(
+ equal(base[0],$s:A),
+ equal(base[1],$s:B),
+ equal(base[2],$s:C),
+ inter(base[0],$s:A), inter(class[0],{subst}),
+ inter(base[1],$s:B), inter(class[1],{interp}),
+ inter(base[2],$s:C), inter(class[2],{ign}),
+ setvar($Pos0, 0),
+ setvar($Pos1, 1),
+ setvar($Pos2, 2),
+ setvar($Head, 0)
+ )
+ </condition>
+
+
+ <instances>
+ <MWE base="all - inclusive">
+ <var name="A">all</var>
+<var name="B">-</var>
+<var name="C">inclusive</var>
+ </MWE>
+ </instances>
+</mwegroup>
+ <mwegroup class="neo" id="Prep" name="Prep" type="fix">
+
+ <condition>
+ and(
+ equal(base[0],$s:A),
+ inter(base[0],$s:A), inter(class[0],{prep}),
+ setvar($Pos0, 0),
+ setvar($Head, 0)
+ )
+ </condition>
+
+
+ <instances>
+
+ <MWE base="ZO">
+ <var name="A">z</var>
+ </MWE>
+
+ </instances>
+
+ </mwegroup>
+ </units_description>
diff --git a/src/tests/python-tests/testdata/wccl_dict_03.xml b/src/tests/python-tests/testdata/wccl_dict_03.xml
new file mode 100644
index 0000000000000000000000000000000000000000..32ca85416a393998822300b3987993c238fa1737
--- /dev/null
+++ b/src/tests/python-tests/testdata/wccl_dict_03.xml
@@ -0,0 +1,217 @@
+<units_description tagset="nkjp">
+
+ <mwegroup class="neo" id="SubstAdj" name="SubstAdj" type="fix">
+
+ <condition>
+ and(
+ equal(base[0],$s:A),
+ equal(base[1],$s:B),
+ inter(base[0],$s:A), inter(class[0],{subst}),
+ inter(base[1],$s:B), inter(class[1],{adj}),
+ setvar($Pos0, 0),
+ setvar($Pos1, 1),
+ setvar($Head, 0)
+ )
+ </condition>
+
+
+ <instances>
+
+ <MWE base="tenis stołowy">
+ <var name="A">tenis</var>
+<var name="B">stołowy</var>
+ </MWE>
+
+ </instances>
+
+ </mwegroup>
+
+ <mwegroup class="neo" id="SubstSubst" name="SubstSubst" type="fix">
+
+ <condition>
+ and(
+ equal(base[0],$s:A),
+ equal(base[1],$s:B),
+ inter(base[0],$s:A), inter(class[0],{subst}),
+ inter(base[1],$s:B), inter(class[1],{subst}),
+ setvar($Pos0, 0),
+ setvar($Pos1, 1),
+ setvar($Head, 0)
+ )
+ </condition>
+
+
+ <instances>
+
+ <MWE base="plac zabaw">
+ <var name="A">plac</var>
+<var name="B">zabawa</var>
+ </MWE>
+
+ </instances>
+
+ </mwegroup>
+
+ <mwegroup class="neo" id="Subst" name="Subst" type="fix">
+
+ <condition>
+ and(
+ equal(base[0],$s:A),
+ inter(base[0],$s:A), inter(class[0],{subst}),
+ setvar($Pos0, 0),
+ setvar($Head, 0)
+ )
+ </condition>
+
+
+ <instances>
+
+ <MWE base="tenis">
+ <var name="A">tenis</var>
+ </MWE>
+
+
+ <MWE base="plac">
+ <var name="A">plac</var>
+ </MWE>
+
+ </instances>
+
+ </mwegroup>
+
+ <mwegroup class="neo" id="SubstIgn" name="SubstIgn" type="fix">
+
+ <condition>
+ and(
+ equal(base[0],$s:A),
+ equal(base[1],$s:B),
+ inter(base[0],$s:A), inter(class[0],{subst}),
+ inter(base[1],$s:B), inter(class[1],{adj}),
+ setvar($Pos0, 0),
+ setvar($Pos1, 1),
+ setvar($Head, 0)
+ )
+ </condition>
+
+
+ <instances>
+ <MWE base="all inclusive">
+ <var name="A">all</var>
+<var name="B">inclusive</var>
+ </MWE>
+
+ </instances>
+
+ </mwegroup>
+
+
+ <mwegroup class="neo" id="Subst" name="Subst" type="fix">
+
+ <condition>
+ and(
+ equal(base[0],$s:A),
+ inter(base[0],$s:A), inter(class[0],{subst}),
+ setvar($Pos0, 0),
+ setvar($Head, 0)
+ )
+ </condition>
+
+
+ <instances>
+
+ <MWE base="tenis">
+ <var name="A">tenis</var>
+ </MWE>
+
+
+ <MWE base="plac">
+ <var name="A">plac</var>
+ </MWE>
+
+ </instances>
+
+ </mwegroup>
+
+
+ <mwegroup class="neo" id="QubSubstInterpIgn" name="QubSubstInterpIgn" type="fix">
+
+ <condition>
+ and(
+ equal(base[0],$s:A),
+ equal(base[1],$s:B),
+ equal(base[2],$s:C),
+ equal(base[3],$s:D),
+ inter(base[0],$s:A), inter(class[0],{qub}),
+ inter(base[1],$s:B), inter(class[1],{subst}),
+ inter(base[2],$s:C), inter(class[2],{interp}),
+ inter(base[3],$s:D), inter(class[3],{ign}),
+ setvar($Pos0, 0),
+ setvar($Pos1, 1),
+ setvar($Pos2, 2),
+ setvar($Pos3, 3),
+ setvar($Head, 0)
+ )
+ </condition>
+
+
+ <instances>
+
+ <MWE base="ultra all - inclusive">
+ <var name="A">ultra</var>
+<var name="B">all</var>
+<var name="C">-</var>
+<var name="D">inclusive</var>
+ </MWE>
+
+ </instances>
+
+ </mwegroup>
+
+ <mwegroup class="neo" id="SubstInterpIgn" name="SubstInterpIgn" type="fix">
+
+ <condition>
+ and(
+ equal(base[0],$s:A),
+ equal(base[1],$s:B),
+ equal(base[2],$s:C),
+ inter(base[0],$s:A), inter(class[0],{subst}),
+ inter(base[1],$s:B), inter(class[1],{interp}),
+ inter(base[2],$s:C), inter(class[2],{ign}),
+ setvar($Pos0, 0),
+ setvar($Pos1, 1),
+ setvar($Pos2, 2),
+ setvar($Head, 0)
+ )
+ </condition>
+
+
+ <instances>
+ <MWE base="all - inclusive">
+ <var name="A">all</var>
+<var name="B">-</var>
+<var name="C">inclusive</var>
+ </MWE>
+ </instances>
+</mwegroup>
+ <mwegroup class="neo" id="Prep" name="Prep" type="fix">
+
+ <condition>
+ and(
+ equal(base[0],$s:A),
+ inter(base[0],$s:A), inter(class[0],{prep}),
+ setvar($Pos0, 0),
+ setvar($Head, 0)
+ )
+ </condition>
+
+
+ <instances>
+
+ <MWE base="ZO">
+ <var name="A">z</var>
+ </MWE>
+
+ </instances>
+
+ </mwegroup>
+ </units_description>
diff --git a/src/tests/python-tests/testdata/wccl_dict_03a.xml b/src/tests/python-tests/testdata/wccl_dict_03a.xml
new file mode 100644
index 0000000000000000000000000000000000000000..9cecfb3aacdd44db5d350d7637cb17170165fd4d
--- /dev/null
+++ b/src/tests/python-tests/testdata/wccl_dict_03a.xml
@@ -0,0 +1,56 @@
+<units_description tagset="nkjp">
+
+ <mwegroup class="neo" id="SubstSubst" name="SubstSubst" type="fix">
+
+ <condition>
+ and(
+ equal(base[0],$s:A),
+ equal(base[1],$s:B),
+ inter(base[0],$s:A), inter(class[0],{subst}),
+ inter(base[1],$s:B), inter(class[1],{subst}),
+ setvar($Pos0, 0),
+ setvar($Pos1, 1),
+ setvar($Head, 0)
+ )
+ </condition>
+
+
+ <instances>
+
+ <MWE base="plac zabaw">
+ <var name="A">plac</var>
+<var name="B">zabawa</var>
+ </MWE>
+
+ </instances>
+
+ </mwegroup>
+
+ <mwegroup class="neo" id="Subst" name="Subst" type="fix">
+
+ <condition>
+ and(
+ equal(base[0],$s:A),
+ inter(base[0],$s:A), inter(class[0],{subst}),
+ setvar($Pos0, 0),
+ setvar($Head, 0)
+ )
+ </condition>
+
+
+ <instances>
+
+ <MWE base="tenis">
+ <var name="A">tenis</var>
+ </MWE>
+
+
+ <MWE base="plac">
+ <var name="A">plac</var>
+ </MWE>
+
+ </instances>
+
+ </mwegroup>
+
+ </units_description>
diff --git a/src/tests/python-tests/testdata/wccl_dict_03b.xml b/src/tests/python-tests/testdata/wccl_dict_03b.xml
new file mode 100644
index 0000000000000000000000000000000000000000..4a671bd890c736346427a248f6e56b99d8fda1ec
--- /dev/null
+++ b/src/tests/python-tests/testdata/wccl_dict_03b.xml
@@ -0,0 +1,56 @@
+<units_description tagset="nkjp">
+
+ <mwegroup class="neo" id="Subst" name="Subst" type="fix">
+
+ <condition>
+ and(
+ equal(base[0],$s:A),
+ inter(base[0],$s:A), inter(class[0],{subst}),
+ setvar($Pos0, 0),
+ setvar($Head, 0)
+ )
+ </condition>
+
+
+ <instances>
+
+ <MWE base="tenis">
+ <var name="A">tenis</var>
+ </MWE>
+
+
+ <MWE base="plac">
+ <var name="A">plac</var>
+ </MWE>
+
+ </instances>
+
+ </mwegroup>
+
+ <mwegroup class="neo" id="SubstSubst" name="SubstSubst" type="fix">
+
+ <condition>
+ and(
+ equal(base[0],$s:A),
+ equal(base[1],$s:B),
+ inter(base[0],$s:A), inter(class[0],{subst}),
+ inter(base[1],$s:B), inter(class[1],{subst}),
+ setvar($Pos0, 0),
+ setvar($Pos1, 1),
+ setvar($Head, 0)
+ )
+ </condition>
+
+
+ <instances>
+
+ <MWE base="plac zabaw">
+ <var name="A">plac</var>
+<var name="B">zabawa</var>
+ </MWE>
+
+ </instances>
+
+ </mwegroup>
+
+ </units_description>
diff --git a/src/tests/python-tests/testdata/wccl_dict_04a.xml b/src/tests/python-tests/testdata/wccl_dict_04a.xml
new file mode 100644
index 0000000000000000000000000000000000000000..b1685d740b9d819a2c0e195a5d36f99cab8c3d47
--- /dev/null
+++ b/src/tests/python-tests/testdata/wccl_dict_04a.xml
@@ -0,0 +1,54 @@
+<units_description tagset="nkjp">
+
+ <mwegroup class="neo" id="SubstSubst" name="SubstSubst" type="fix">
+
+ <condition>
+ and(
+ equal(base[0],$s:A),
+ equal(base[1],$s:B),
+ inter(base[0],$s:A), inter(class[0],{subst}),
+ inter(base[1],$s:B), inter(class[1],{subst}),
+ setvar($Pos0, 0),
+ setvar($Pos1, 1),
+ setvar($Head, 0)
+ )
+ </condition>
+
+
+ <instances>
+
+ <MWE base="plac zabaw">
+ <var name="A">plac</var>
+<var name="B">zabawa</var>
+ </MWE>
+
+ </instances>
+
+ </mwegroup>
+
+ <mwegroup class="neo" id="Subst" name="Subst" type="fix">
+
+ <condition>
+ and(
+ equal(base[0],$s:A),
+ inter(base[0],$s:A), inter(class[0],{subst}),
+ setvar($Pos0, 0),
+ setvar($Head, 0)
+ )
+ </condition>
+
+
+ <instances>
+
+ <MWE base="plac">
+ <var name="A">plac</var>
+ </MWE>
+
+ <MWE base="zabawa">
+ <var name="A">zabawa</var>
+ </MWE>
+ </instances>
+
+ </mwegroup>
+
+ </units_description>
diff --git a/src/tests/python-tests/testdata/wccl_dict_04b.xml b/src/tests/python-tests/testdata/wccl_dict_04b.xml
new file mode 100644
index 0000000000000000000000000000000000000000..8015ddbde0ecd3fa0f990242fe17d059376ea752
--- /dev/null
+++ b/src/tests/python-tests/testdata/wccl_dict_04b.xml
@@ -0,0 +1,53 @@
+<units_description tagset="nkjp">
+ <mwegroup class="neo" id="Subst" name="Subst" type="fix">
+
+ <condition>
+ and(
+ equal(base[0],$s:A),
+ inter(base[0],$s:A), inter(class[0],{subst}),
+ setvar($Pos0, 0),
+ setvar($Head, 0)
+ )
+ </condition>
+
+
+ <instances>
+
+ <MWE base="plac">
+ <var name="A">plac</var>
+ </MWE>
+
+ <MWE base="zabawa">
+ <var name="A">zabawa</var>
+ </MWE>
+ </instances>
+
+ </mwegroup>
+
+ <mwegroup class="neo" id="SubstSubst" name="SubstSubst" type="fix">
+
+ <condition>
+ and(
+ equal(base[0],$s:A),
+ equal(base[1],$s:B),
+ inter(base[0],$s:A), inter(class[0],{subst}),
+ inter(base[1],$s:B), inter(class[1],{subst}),
+ setvar($Pos0, 0),
+ setvar($Pos1, 1),
+ setvar($Head, 0)
+ )
+ </condition>
+
+
+ <instances>
+
+ <MWE base="plac zabaw">
+ <var name="A">plac</var>
+<var name="B">zabawa</var>
+ </MWE>
+
+ </instances>
+
+ </mwegroup>
+
+ </units_description>
diff --git a/src/tests/python-tests/testdata/wccl_dict_04c.xml b/src/tests/python-tests/testdata/wccl_dict_04c.xml
new file mode 100644
index 0000000000000000000000000000000000000000..e3cd50d96375a4f4cb329815dd0e982314ad1528
--- /dev/null
+++ b/src/tests/python-tests/testdata/wccl_dict_04c.xml
@@ -0,0 +1,53 @@
+<units_description tagset="nkjp">
+ <mwegroup class="neo" id="Subst" name="Subst" type="fix">
+
+ <condition>
+ and(
+ equal(base[0],$s:A),
+ inter(base[0],$s:A), inter(class[0],{subst}),
+ setvar($Pos0, 0),
+ setvar($Head, 0)
+ )
+ </condition>
+
+
+ <instances>
+
+ <MWE base="plac">
+ <var name="A">plac</var>
+ </MWE>
+
+ <MWE base="zabawa">
+ <var name="A">zabawa</var>
+ </MWE>
+ </instances>
+
+ </mwegroup>
+
+ <mwegroup class="neo" id="SubstSubst" name="SubstSubst" type="fix">
+
+ <condition>
+ and(
+ equal(base[0],$s:A),
+ equal(base[1],$s:B),
+ inter(base[0],$s:A), inter(class[0],{subst}),
+ inter(base[1],$s:B), inter(class[1],{subst}),
+ setvar($Pos0, 0),
+ setvar($Pos1, 1),
+ setvar($Head, 1)
+ )
+ </condition>
+
+
+ <instances>
+
+ <MWE base="plac zabaw">
+ <var name="A">plac</var>
+<var name="B">zabawa</var>
+ </MWE>
+
+ </instances>
+
+ </mwegroup>
+
+ </units_description>