diff --git a/libwccl/CMakeLists.txt b/libwccl/CMakeLists.txt
index 7c77e0a346869b6b5e5d462582f78d203473939a..a763a8f4b7241e647a0216c7d0ed1daa91aacc75 100644
--- a/libwccl/CMakeLists.txt
+++ b/libwccl/CMakeLists.txt
@@ -67,7 +67,9 @@ SET(libwccl_STAT_SRC
ops/functions/tset/catfilter.cpp
ops/functions/tset/getsymbols.cpp
ops/functions/tset/getsymbolsinrange.cpp
+ ops/match/actions/markbase.cpp
ops/match/actions/markmatch.cpp
+ ops/match/actions/overwritematch.cpp
ops/match/actions/unmarkmatch.cpp
ops/match/applyoperator.cpp
ops/match/conditions/conjconditions.cpp
diff --git a/libwccl/ops/match/actions/markbase.cpp b/libwccl/ops/match/actions/markbase.cpp
new file mode 100644
index 0000000000000000000000000000000000000000..f38d9e32c12df856c0420291eaa4bfd147f4ec20
--- /dev/null
+++ b/libwccl/ops/match/actions/markbase.cpp
@@ -0,0 +1,109 @@
+#include <libwccl/values/match.h>
+#include <libwccl/ops/match/actions/markmatch.h>
+
+#include <sstream>
+
+
+namespace Wccl {
+namespace Matching {
+
+void MarkBase::execute_mark(const ActionExecContext& context,
+ bool may_overwrite) const
+{
+ // TODO may overwrite, use it in mark_mark and perhaps move fields back to private
+ SentenceContext& sc = context.sentence_context();
+ boost::shared_ptr<Corpus2::AnnotatedSentence> as;
+ as = boost::dynamic_pointer_cast<Corpus2::AnnotatedSentence>(sc.get_sentence_ptr());
+ if (!as) {
+ throw InvalidArgument("context", "Operator needs an annotated sentence.");
+ }
+
+ boost::shared_ptr<const Match> match_from = match_from_->apply(context);
+ boost::shared_ptr<const Match> match_to =
+ (match_from_ == match_to_) ? match_from : match_to_->apply(context);
+ boost::shared_ptr<const Match> head_match =
+ (match_from_ == head_match_) ? match_from : head_match_->apply(context);
+
+ int abs_left = match_from->first_token(as);
+ if (abs_left < 0) {
+ throw WcclError("Received starting match that points outside sentence.");
+ }
+
+ int abs_right = match_to->last_token(as);
+ if (abs_right >= sc.size()) {
+ throw WcclError("Received ending match that points outside sentence.");
+ }
+ if (abs_left > abs_right) {
+ throw WcclError("Received starting match points after the received ending match.");
+ }
+
+ int abs_head = head_match->first_token(as);
+ if (abs_head < abs_left || abs_head > abs_right) {
+ throw WcclError("Received head match points outside range defined by start and end matches.");
+ }
+
+ if (!as->has_channel(chan_name_)) {
+ as->create_channel(chan_name_);
+ }
+ Corpus2::AnnotationChannel& channel = as->get_channel(chan_name_);
+
+ int segment_idx = channel.get_new_segment_index();
+
+ if (may_overwrite) {
+ std::set<int> segments_to_unmark;
+ for (int i = abs_left; i <= abs_right; ++i) {
+ if (channel.get_segment_at(i) > 0) {
+ segments_to_unmark.insert(channel.get_segment_at(i));
+ }
+ }
+ if (!segments_to_unmark.empty()) {
+ // unmark all the segments that appeared within the range
+ for (int i = 0; i < channel.size(); ++i) {
+ if (segments_to_unmark.find(channel.segments()[i]) !=
+ segments_to_unmark.end()) {
+ channel.set_segment_at(i, 0);
+ channel.set_head_at(i, false);
+ }
+ }
+ }
+ }
+ else {
+ for (int i = abs_left; i <= abs_right; ++i) {
+ if (channel.get_segment_at(i) > 0) {
+ throw WcclError("Mark action would overwrite existing annotation");
+ }
+ }
+ }
+ // mark new one
+ for (int i = abs_left; i <= abs_right; ++i) {
+ channel.set_segment_at(i, segment_idx);
+ channel.set_head_at(i, false);
+ }
+ channel.set_head_at(abs_head, true);
+}
+
+std::string MarkBase::to_string(const Corpus2::Tagset& tagset) const
+{
+ std::ostringstream os;
+ os << name() << "("
+ << match_from_->to_string(tagset) << ", ";
+ if (match_from_.get() != match_to_.get()) {
+ os << match_to_->to_string(tagset) << ", ";
+ }
+ os << "\"" << chan_name_ << "\")";
+ return os.str();
+}
+
+std::ostream& MarkBase::write_to(std::ostream& os) const
+{
+ os << name() << "("
+ << *match_from_ << ", ";
+ if (match_from_.get() != match_to_.get()) {
+ os << *match_to_ << ", ";
+ }
+ os << "\"" << chan_name_ << "\")";
+ return os;
+}
+
+} /* end ns Matching */
+} /* end ns Wccl */
diff --git a/libwccl/ops/match/actions/markbase.h b/libwccl/ops/match/actions/markbase.h
new file mode 100644
index 0000000000000000000000000000000000000000..dfbedf082714eab907203f0ef4d5c8d7a29525b5
--- /dev/null
+++ b/libwccl/ops/match/actions/markbase.h
@@ -0,0 +1,63 @@
+#ifndef LIBWCCL_OPS_MATCH_ACTIONS_MARKBASE_H
+#define LIBWCCL_OPS_MATCH_ACTIONS_MARKBASE_H
+
+
+#include <libwccl/ops/match/matchaction.h>
+#include <libwccl/ops/function.h>
+
+namespace Wccl {
+namespace Matching {
+
+class MarkBase : public MatchAction
+{
+public:
+ MarkBase(
+ const boost::shared_ptr<Function<Match> >& match_from,
+ const boost::shared_ptr<Function<Match> >& match_to,
+ const boost::shared_ptr<Function<Match> >& head_match,
+ const std::string& annotation_name)
+ : match_from_(match_from),
+ match_to_(match_to),
+ head_match_(head_match),
+ chan_name_(annotation_name)
+ {
+ BOOST_ASSERT(match_from_);
+ BOOST_ASSERT(match_to_);
+ BOOST_ASSERT(head_match_);
+ }
+
+ virtual std::string name() const = 0;
+
+ /**
+ * Executes the action for the given execution context.
+ */
+ virtual void execute(const ActionExecContext& context) const = 0;
+
+ std::string to_string(const Corpus2::Tagset& tagset) const;
+
+protected:
+ const boost::shared_ptr<Function<Match> > match_from_;
+ const boost::shared_ptr<Function<Match> > match_to_;
+ const boost::shared_ptr<Function<Match> > head_match_;
+ const std::string chan_name_;
+
+ /**
+ * Writes string representation of the MatchAction to
+ * an output stream.
+ * @returns Stream written to.
+ * @note May be incomplete and/or containt internal info.
+ */
+ virtual std::ostream& write_to(std::ostream& ostream) const;
+
+ /**
+ * The actual implementation. Depending on may_overwrite, will either
+ * remove any previous annotations in the range given or will raise
+ * an exception if any annotations present. */
+ void execute_mark(const ActionExecContext& context, bool may_overwrite)
+ const;
+};
+
+} /* end ns Matching */
+} /* end ns Wccl */
+
+#endif // LIBWCCL_OPS_MATCH_ACTIONS_MARKBASE_H
diff --git a/libwccl/ops/match/actions/markmatch.cpp b/libwccl/ops/match/actions/markmatch.cpp
index 2a113abf7de17c32a95ea3227d54e1de5d867775..9674af88ebbcecf3ccbdcac8e8dfb314b7b6e62f 100644
--- a/libwccl/ops/match/actions/markmatch.cpp
+++ b/libwccl/ops/match/actions/markmatch.cpp
@@ -1,4 +1,3 @@
-#include <libwccl/values/match.h>
#include <libwccl/ops/match/actions/markmatch.h>
#include <sstream>
@@ -9,77 +8,8 @@ namespace Matching {
void MarkMatch::execute(const ActionExecContext& context) const
{
- SentenceContext& sc = context.sentence_context();
- boost::shared_ptr<Corpus2::AnnotatedSentence> as;
- as = boost::dynamic_pointer_cast<Corpus2::AnnotatedSentence>(sc.get_sentence_ptr());
- if (!as) {
- throw InvalidArgument("context", "Operator needs an annotated sentence.");
- }
-
- boost::shared_ptr<const Match> match_from = match_from_->apply(context);
- boost::shared_ptr<const Match> match_to =
- (match_from_ == match_to_) ? match_from : match_to_->apply(context);
- boost::shared_ptr<const Match> head_match =
- (match_from_ == head_match_) ? match_from : head_match_->apply(context);
-
- int abs_left = match_from->first_token(as);
- if (abs_left < 0) {
- throw WcclError("Received starting match that points outside sentence.");
- }
-
- int abs_right = match_to->last_token(as);
- if (abs_right >= sc.size()) {
- throw WcclError("Received ending match that points outside sentence.");
- }
- if (abs_left > abs_right) {
- throw WcclError("Received starting match points after the received ending match.");
- }
-
- int abs_head = head_match->first_token(as);
- if (abs_head < abs_left || abs_head > abs_right) {
- throw WcclError("Received head match points outside range defined by start and end matches.");
- }
-
- if (!as->has_channel(chan_name_)) {
- as->create_channel(chan_name_);
- }
- Corpus2::AnnotationChannel& channel = as->get_channel(chan_name_);
-
- int segment_idx = channel.get_new_segment_index();
-
- for (int i = abs_left; i <= abs_right; ++i) {
- if (channel.get_segment_at(i) > 0) {
- throw WcclError("Mark action would overwrite existing annotation");
- }
- }
- for (int i = abs_left; i <= abs_right; ++i) {
- channel.set_segment_at(i, segment_idx);
- channel.set_head_at(i, false);
- }
- channel.set_head_at(abs_head, true);
-}
-
-std::string MarkMatch::to_string(const Corpus2::Tagset& tagset) const
-{
- std::ostringstream os;
- os << name() << "("
- << match_from_->to_string(tagset) << ", ";
- if (match_from_.get() != match_to_.get()) {
- os << match_to_->to_string(tagset) << ", ";
- }
- os << "\"" << chan_name_ << "\")";
- return os.str();
-}
-
-std::ostream& MarkMatch::write_to(std::ostream& os) const
-{
- os << name() << "("
- << *match_from_ << ", ";
- if (match_from_.get() != match_to_.get()) {
- os << *match_to_ << ", ";
- }
- os << "\"" << chan_name_ << "\")";
- return os;
+ // fire inner execute in fail-on-annotation-overwrite mode
+ execute_mark(context, false);
}
} /* end ns Matching */
diff --git a/libwccl/ops/match/actions/markmatch.h b/libwccl/ops/match/actions/markmatch.h
index 514ec19e8c05358068a906a5b247424567d7994a..3f1cd3d8721c04afb2abeec099e71cab3719a532 100644
--- a/libwccl/ops/match/actions/markmatch.h
+++ b/libwccl/ops/match/actions/markmatch.h
@@ -1,13 +1,12 @@
#ifndef LIBWCCL_OPS_MATCH_ACTIONS_MARKMATCH_H
#define LIBWCCL_OPS_MATCH_ACTIONS_MARKMATCH_H
-#include <libwccl/ops/match/matchaction.h>
-#include <libwccl/ops/function.h>
+#include <libwccl/ops/match/actions/markbase.h>
namespace Wccl {
namespace Matching {
-class MarkMatch : public MatchAction
+class MarkMatch : public MarkBase
{
public:
MarkMatch(
@@ -15,42 +14,25 @@ public:
const boost::shared_ptr<Function<Match> >& match_to,
const boost::shared_ptr<Function<Match> >& head_match,
const std::string& annotation_name)
- : match_from_(match_from),
- match_to_(match_to),
- head_match_(head_match),
- chan_name_(annotation_name)
+ : MarkBase(match_from, match_to, head_match, annotation_name)
{
- BOOST_ASSERT(match_from_);
- BOOST_ASSERT(match_to_);
- BOOST_ASSERT(head_match_);
}
MarkMatch(
const boost::shared_ptr<Function<Match> >& match_from,
const boost::shared_ptr<Function<Match> >& match_to,
const std::string& annotation_name)
- : match_from_(match_from),
- match_to_(match_to),
- head_match_(match_from),
- chan_name_(annotation_name)
+ : MarkBase(match_from, match_to, match_from, annotation_name)
{
- BOOST_ASSERT(match_from_);
- BOOST_ASSERT(match_to_);
- BOOST_ASSERT(head_match_);
}
MarkMatch(
const boost::shared_ptr<Function<Match> >& match_from_to,
const std::string& annotation_name)
- : match_from_(match_from_to),
- match_to_(match_from_to),
- head_match_(match_from_to),
- chan_name_(annotation_name)
+ : MarkBase(match_from_to, match_from_to, match_from_to, annotation_name)
{
- BOOST_ASSERT(match_from_);
- BOOST_ASSERT(match_to_);
- BOOST_ASSERT(head_match_);
}
+
/**
* @returns Name of the action.
*/
@@ -62,25 +44,6 @@ public:
* Executes the action for the given execution context.
*/
void execute(const ActionExecContext& context) const;
-
- /**
- * @returns String representation of the expression.
- */
- std::string to_string(const Corpus2::Tagset& tagset) const;
-
-protected:
- /**
- * Writes string representation of the MatchAction to
- * an output stream.
- * @returns Stream written to.
- * @note May be incomplete and/or containt internal info.
- */
- virtual std::ostream& write_to(std::ostream& ostream) const;
-private:
- const boost::shared_ptr<Function<Match> > match_from_;
- const boost::shared_ptr<Function<Match> > match_to_;
- const boost::shared_ptr<Function<Match> > head_match_;
- const std::string chan_name_;
};
} /* end ns Matching */
diff --git a/libwccl/ops/match/actions/overwritematch.cpp b/libwccl/ops/match/actions/overwritematch.cpp
new file mode 100644
index 0000000000000000000000000000000000000000..e8fafa7bef924dd6aadced33d5a56332d2f17fe9
--- /dev/null
+++ b/libwccl/ops/match/actions/overwritematch.cpp
@@ -0,0 +1,16 @@
+#include <libwccl/ops/match/actions/overwritematch.h>
+
+#include <sstream>
+
+
+namespace Wccl {
+namespace Matching {
+
+void OverwriteMatch::execute(const ActionExecContext& context) const
+{
+ // fire inner execute in silent-overwrite mode
+ execute_mark(context, true);
+}
+
+} /* end ns Matching */
+} /* end ns Wccl */
diff --git a/libwccl/ops/match/actions/overwritematch.h b/libwccl/ops/match/actions/overwritematch.h
new file mode 100644
index 0000000000000000000000000000000000000000..87dbd2e73e17384c68df851e511e3e1709bb13fe
--- /dev/null
+++ b/libwccl/ops/match/actions/overwritematch.h
@@ -0,0 +1,52 @@
+#ifndef LIBWCCL_OPS_MATCH_ACTIONS_OVERWRITEMATCH_H
+#define LIBWCCL_OPS_MATCH_ACTIONS_OVERWRITEMATCH_H
+
+#include <libwccl/ops/match/actions/markbase.h>
+
+namespace Wccl {
+namespace Matching {
+
+class OverwriteMatch : public MarkBase
+{
+public:
+ OverwriteMatch(
+ const boost::shared_ptr<Function<Match> >& match_from,
+ const boost::shared_ptr<Function<Match> >& match_to,
+ const boost::shared_ptr<Function<Match> >& head_match,
+ const std::string& annotation_name)
+ : MarkBase(match_from, match_to, head_match, annotation_name)
+ {
+ }
+
+ OverwriteMatch(
+ const boost::shared_ptr<Function<Match> >& match_from,
+ const boost::shared_ptr<Function<Match> >& match_to,
+ const std::string& annotation_name)
+ : MarkBase(match_from, match_to, match_from, annotation_name)
+ {
+ }
+
+ OverwriteMatch(
+ const boost::shared_ptr<Function<Match> >& match_from_to,
+ const std::string& annotation_name)
+ : MarkBase(match_from_to, match_from_to, match_from_to, annotation_name)
+ {
+ }
+
+ /**
+ * @returns Name of the action.
+ */
+ std::string name() const {
+ return "remark";
+ }
+
+ /**
+ * Executes the action for the given execution context.
+ */
+ void execute(const ActionExecContext& context) const;
+};
+
+} /* end ns Matching */
+} /* end ns Wccl */
+
+#endif // LIBWCCL_OPS_MATCH_ACTIONS_OVERWRITEMATCH_H
diff --git a/libwccl/parser/grammar.g b/libwccl/parser/grammar.g
index 5d6e6db82796dda93b800cee6efa802d3f04d0ff..7aa71ed78c8738b6656bcdf1c0f2e8a545564a41 100644
--- a/libwccl/parser/grammar.g
+++ b/libwccl/parser/grammar.g
@@ -97,6 +97,7 @@ header {
#include <libwccl/ops/match/conditions/longest.h>
#include <libwccl/ops/match/actions/markmatch.h>
#include <libwccl/ops/match/actions/unmarkmatch.h>
+ #include <libwccl/ops/match/actions/overwritematch.h>
#include <libwccl/ops/functions/match/submatch.h>
// Wccl whole file syntax
@@ -2686,6 +2687,7 @@ match_action
returns [boost::shared_ptr<Matching::MatchAction> m_act]
: m_act = match_mark_action [scope]
| m_act = match_unmark_action [scope]
+ | m_act = match_remark_action [scope]
;
// Match mark action
@@ -2748,6 +2750,47 @@ match_unmark_action
}
;
+// Overwrite-match mark action ("remark")
+// Returns
+match_remark_action
+ [ParsingScope& scope]
+ returns [boost::shared_ptr<Matching::OverwriteMatch> m_act]
+{
+ boost::shared_ptr<Function<Match> > match_to;
+ boost::shared_ptr<Function<Match> > match_from;
+ boost::shared_ptr<Function<Match> > head_match;
+}
+ : "remark" LPAREN
+ match_from = match_operator[scope] COMMA
+ ( match_to = match_operator[scope] COMMA
+ ( head_match = match_operator[scope] COMMA )?
+ )?
+ annotation_name : STRING
+ RPAREN {
+ if (!match_to) {
+ m_act.reset(
+ new Matching::OverwriteMatch(
+ match_from,
+ ((antlr::Token*)annotation_name)->getText()));
+ } else {
+ if (!head_match) {
+ m_act.reset(
+ new Matching::OverwriteMatch(
+ match_from,
+ match_to,
+ ((antlr::Token*)annotation_name)->getText()));
+ } else {
+ m_act.reset(
+ new Matching::OverwriteMatch(
+ match_from,
+ match_to,
+ head_match,
+ ((antlr::Token*)annotation_name)->getText()));
+ }
+ }
+ }
+;
+
// Match action separated by comma
// Returns boost::shared_ptr<std::vector<boost::shared_ptr<Matching::MatchAction> > >
match_action_comma_sep
diff --git a/tests/rules-data/match/remark/cclmatch.xml b/tests/rules-data/match/remark/cclmatch.xml
new file mode 100644
index 0000000000000000000000000000000000000000..25498def4c4f08d429448416989c31e7e70274a0
--- /dev/null
+++ b/tests/rules-data/match/remark/cclmatch.xml
@@ -0,0 +1,90 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE cesAna SYSTEM "xcesAnaIPI.dtd">
+<chunkList xmlns:xlink="http://www.w3.org/1999/xlink">
+ <chunk>
+ <sentence>
+ <tok>
+ <orth>Miasto</orth>
+ <lex><base>miasto</base><ctag>noun</ctag></lex>
+ <ann chan="AdjN">0</ann>
+ <ann chan="P" head="1">1</ann>
+ </tok>
+ <tok>
+ <orth>stołeczne</orth>
+ <lex><base>stołeczny</base><ctag>adj</ctag></lex>
+ <ann chan="AdjN" head="1">1</ann>
+ <ann chan="P">0</ann>
+ </tok>
+ <tok>
+ <orth>Warszawa</orth>
+ <lex><base>warszawa</base><ctag>noun</ctag></lex>
+ <ann chan="AdjN">1</ann>
+ <ann chan="P">0</ann>
+ </tok>
+ <tok>
+ <orth>centralna</orth>
+ <lex><base>centralny</base><ctag>adj</ctag></lex>
+ <ann chan="AdjN">0</ann>
+ <ann chan="P">0</ann>
+ </tok>
+ <ns/>
+ <tok>
+ <orth>,</orth>
+ <lex><base>,</base><ctag>interp</ctag></lex>
+ <ann chan="AdjN">0</ann>
+ <ann chan="P">0</ann>
+ </tok>
+ <tok>
+ <orth>nowy</orth>
+ <lex><base>nowy</base><ctag>adj</ctag></lex>
+ <ann chan="AdjN">0</ann>
+ <ann chan="P" head="1">2</ann>
+ </tok>
+ <tok>
+ <orth>wielki</orth>
+ <lex><base>wielki</base><ctag>adj</ctag></lex>
+ <ann chan="AdjN" head="1">2</ann>
+ <ann chan="P">0</ann>
+ </tok>
+ <tok>
+ <orth>Berlin</orth>
+ <lex><base>berlin</base><ctag>noun</ctag></lex>
+ <ann chan="AdjN">2</ann>
+ <ann chan="P">0</ann>
+ </tok>
+ <tok>
+ <orth>Zachodni</orth>
+ <lex><base>zachodni</base><ctag>adj</ctag></lex>
+ <lex><base>zachodny</base><ctag>adj</ctag></lex>
+ <ann chan="AdjN">0</ann>
+ <ann chan="P">0</ann>
+ </tok>
+ <tok>
+ <orth>z</orth>
+ <lex><base>z</base><ctag>other</ctag></lex>
+ <lex><base>z</base><ctag>qub</ctag></lex>
+ <ann chan="AdjN">0</ann>
+ <ann chan="P">0</ann>
+ </tok>
+ <tok>
+ <orth>dzielnicÄ…</orth>
+ <lex><base>dzielnica</base><ctag>noun</ctag></lex>
+ <ann chan="AdjN">0</ann>
+ <ann chan="P">0</ann>
+ </tok>
+ <tok>
+ <orth>tureckÄ…</orth>
+ <lex><base>turecki</base><ctag>adj</ctag></lex>
+ <ann chan="AdjN">0</ann>
+ <ann chan="P">0</ann>
+ </tok>
+ <ns/>
+ <tok>
+ <orth>.</orth>
+ <lex><base>.</base><ctag>interp</ctag></lex>
+ <ann chan="AdjN">0</ann>
+ <ann chan="P">0</ann>
+ </tok>
+ </sentence>
+ </chunk>
+</chunkList>
diff --git a/tests/rules-data/match/remark/cclmatch1.ccl b/tests/rules-data/match/remark/cclmatch1.ccl
new file mode 100644
index 0000000000000000000000000000000000000000..a207bc81a94c82b31fd0aa77de3f57c199e37826
--- /dev/null
+++ b/tests/rules-data/match/remark/cclmatch1.ccl
@@ -0,0 +1,25 @@
+match_rules(
+ apply(
+ match(
+ is("P"),
+ inter(class[0], adj)
+ ),
+ actions(
+ remark( M, "P" )
+ )
+ ) ;
+
+
+ apply(
+ match(
+ inter(base[0], ["zachodni"]),
+ inter(base[0], ["z"]),
+ inter(base[0], ["dzielnica"]),
+ inter(base[0], ["turecki"])
+ ),
+ actions(
+ remark( M, "AdjN" )
+ )
+ )
+
+)
diff --git a/tests/rules-data/match/remark/cclmatch1.out.xml b/tests/rules-data/match/remark/cclmatch1.out.xml
new file mode 100644
index 0000000000000000000000000000000000000000..2f389b0cb810b8d9b1c08bd05161b3868415ba1c
--- /dev/null
+++ b/tests/rules-data/match/remark/cclmatch1.out.xml
@@ -0,0 +1,90 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE cesAna SYSTEM "xcesAnaIPI.dtd">
+<chunkList xmlns:xlink="http://www.w3.org/1999/xlink">
+ <chunk>
+ <sentence>
+ <tok>
+ <orth>Miasto</orth>
+ <lex><base>miasto</base><ctag>noun</ctag></lex>
+ <ann chan="AdjN">0</ann>
+ <ann chan="P" head="1">3</ann>
+ </tok>
+ <tok>
+ <orth>stołeczne</orth>
+ <lex><base>stołeczny</base><ctag>adj</ctag></lex>
+ <ann chan="AdjN" head="1">1</ann>
+ <ann chan="P">3</ann>
+ </tok>
+ <tok>
+ <orth>Warszawa</orth>
+ <lex><base>warszawa</base><ctag>noun</ctag></lex>
+ <ann chan="AdjN">1</ann>
+ <ann chan="P">0</ann>
+ </tok>
+ <tok>
+ <orth>centralna</orth>
+ <lex><base>centralny</base><ctag>adj</ctag></lex>
+ <ann chan="AdjN">0</ann>
+ <ann chan="P">0</ann>
+ </tok>
+ <ns/>
+ <tok>
+ <orth>,</orth>
+ <lex><base>,</base><ctag>interp</ctag></lex>
+ <ann chan="AdjN">0</ann>
+ <ann chan="P">0</ann>
+ </tok>
+ <tok>
+ <orth>nowy</orth>
+ <lex><base>nowy</base><ctag>adj</ctag></lex>
+ <ann chan="AdjN">0</ann>
+ <ann chan="P" head="1">1</ann>
+ </tok>
+ <tok>
+ <orth>wielki</orth>
+ <lex><base>wielki</base><ctag>adj</ctag></lex>
+ <ann chan="AdjN" head="1">2</ann>
+ <ann chan="P">1</ann>
+ </tok>
+ <tok>
+ <orth>Berlin</orth>
+ <lex><base>berlin</base><ctag>noun</ctag></lex>
+ <ann chan="AdjN">2</ann>
+ <ann chan="P">0</ann>
+ </tok>
+ <tok>
+ <orth>Zachodni</orth>
+ <lex><base>zachodni</base><ctag>adj</ctag></lex>
+ <lex><base>zachodny</base><ctag>adj</ctag></lex>
+ <ann chan="AdjN" head="1">3</ann>
+ <ann chan="P">0</ann>
+ </tok>
+ <tok>
+ <orth>z</orth>
+ <lex><base>z</base><ctag>other</ctag></lex>
+ <lex><base>z</base><ctag>qub</ctag></lex>
+ <ann chan="AdjN">3</ann>
+ <ann chan="P">0</ann>
+ </tok>
+ <tok>
+ <orth>dzielnicÄ…</orth>
+ <lex><base>dzielnica</base><ctag>noun</ctag></lex>
+ <ann chan="AdjN">3</ann>
+ <ann chan="P">0</ann>
+ </tok>
+ <tok>
+ <orth>tureckÄ…</orth>
+ <lex><base>turecki</base><ctag>adj</ctag></lex>
+ <ann chan="AdjN">3</ann>
+ <ann chan="P">0</ann>
+ </tok>
+ <ns/>
+ <tok>
+ <orth>.</orth>
+ <lex><base>.</base><ctag>interp</ctag></lex>
+ <ann chan="AdjN">0</ann>
+ <ann chan="P">0</ann>
+ </tok>
+ </sentence>
+ </chunk>
+</chunkList>
diff --git a/tests/rules-data/match/remark/simple.is-the-tagset b/tests/rules-data/match/remark/simple.is-the-tagset
new file mode 100644
index 0000000000000000000000000000000000000000..8b137891791fe96927ad78e64b0aad7bded08bdc
--- /dev/null
+++ b/tests/rules-data/match/remark/simple.is-the-tagset
@@ -0,0 +1 @@
+