Merge pull request #49 from spraakbanken/44-handle-annotations-when-2…

…-tokens-are-merged

fix: handle annotations when 2 tokens are merged

examples/christoph-borg/config.yaml

@@ -1,16 +1,17 @@
 metadata:
-    id: christoph-borg
-    language: swe
+  id: christoph-borg
+  language: swe
 
 import:
-    importer: text_import:parse
+  importer: text_import:parse
 
 export:
-    annotations:
-        - <sentence>
-        # - <token:word>
-        - <token>:stanza.pos
-        - <token>:sbx_ocr_correction_viklofg_sweocr.ocr-correction--viklofg-sweocr
-
+  annotations:
+    - <sentence>
+    # - <token:word>
+    - <token>:stanza.pos
+    - sbx_ocr_correction_viklofg_sweocr.sbx-ocr-correction
+    - sbx_ocr_correction_viklofg_sweocr.sbx-ocr-correction:sbx_ocr_correction_viklofg_sweocr.ocr-correction--viklofg-sweocr
+
 sparv:
-    compression: none
+  compression: none

examples/ocr-correction-viklofg-sweocr/config.yaml

@@ -1,16 +1,17 @@
 metadata:
-    id: hello-ocr
-    language: swe
+  id: hello-ocr
+  language: swe
 
 import:
-    importer: text_import:parse
+  importer: text_import:parse
 
 export:
-    annotations:
-        - <sentence>
-        # - <token:word>
-        - <token>:stanza.pos
-        - <token>:sbx_ocr_correction_viklofg_sweocr.ocr-correction--viklofg-sweocr
+  annotations:
+    - <sentence>
+    # - <token:word>
+    - <token>:stanza.pos
+    - sbx_ocr_correction_viklofg_sweocr.sbx-ocr-correction
+    - sbx_ocr_correction_viklofg_sweocr.sbx-ocr-correction:sbx_ocr_correction_viklofg_sweocr.ocr-correction--viklofg-sweocr
 
 sparv:
-    compression: none
+  compression: none

examples/texts/dokument.txt

@@ -1 +1,2 @@
-Den i HandelstidniDgens g&rdagsnnmmer omtalade hvalfisken, sorn fångats i Frölnndaviken
+Den i HandelstidniDgens g&rdagsnnmmer omtalade hvalfisken, sorn fångats i Frölnndaviken.
+Jonath an saknades.

ocr-correction-viklofg-sweocr/Makefile

@@ -146,7 +146,7 @@ prepare-release: tests/requirements-testing.lock
 
 # we use lock extension so that dependabot doesn't pick up changes in this file
 tests/requirements-testing.lock: pyproject.toml
-	pdm export --dev --format requirements --output $@
+	pdm export --dev --format requirements --no-hashes --output $@
 
 .PHONY: kb-bert-prepare-release
 kb-bert-prepare-release: ocr-correction-viklofg-sweocr/CHANGELOG.md
@@ -158,4 +158,4 @@ CHANGELOG.md:
 
 .PHONY: ocr-correction-viklofg-sweocr/CHANGELOG.md
 ocr-correction-viklofg-sweocr/CHANGELOG.md:
-	git cliff --unreleased --include-path "ocr-correction-viklofg-sweocr/**/*" --include-path "examples/ocr-correction-viklofg-sweocr/**/*" --prepend $@
+	git cliff --unreleased --include-path "ocr-correction-viklofg-sweocr/**/*" --include-path "examples/ocr-correction-viklofg-sweocr/**/*" --prepend $@