Минимальная конфигурация для Томиты

.gitignore

@@ -0,0 +1,2 @@
+*.bin
+test/data/output/*.xml

extract/Point.cxx

@@ -0,0 +1,5 @@
+#encoding "utf-8"
+
+PointName -> AnyWord<wff=/[A-Z]/>;
+
+Point -> "точка" PointName interp(Point.Name);

extract/Readme.md

@@ -0,0 +1,3 @@
+To get facts from file in `test/data/input` run `extract $file` (on UNIX run `./extract $file`) in **this** directory.
+
+`tomitaparser` **have to be** in PATH (`/usr/bin`)

extract/config.proto

@@ -0,0 +1,19 @@
+encoding "utf-8";
+
+TTextMinerConfig {
+  Dictionary = "mydic.gzt";
+
+  PrettyOutput = "pr.html";
+
+  Articles = [
+    { Name = "точка" }
+  ]
+
+  Facts = [
+    { Name = "Point" }
+  ]
+
+  Output = {
+    Format = xml;
+  }
+}

extract/extract

@@ -0,0 +1 @@
+tomitaparser config.proto < ../test/data/input/$1 > ../test/data/output/$(basename $1 .txt).xml

extract/extract.bat

@@ -0,0 +1,3 @@
+@echo off
+
+tomitaparser config.proto < ..\test\data\input\%1 > ..\test\data\output\%~n1.xml

extract/facttypes.proto

@@ -0,0 +1,6 @@
+import "base.proto";
+import "facttypes_base.proto";
+
+message Point : NFactType.TFact {
+  optional string Name = 1;
+}

extract/mydic.gzt

@@ -0,0 +1,9 @@
+encoding "utf-8";
+
+import "base.proto";
+import "articles_base.proto";
+import "Facttypes.proto";
+
+TAuxDicArticle "точка" {
+  key = { "tomita:Point.cxx" type=CUSTOM }
+}