Merge branch 'master' into fix-polygon-for-parent

OCR-D · Jun 5, 2020 · 0bbd5b5 · 0bbd5b5
2 parents 94719f8 + bfd29aa
commit 0bbd5b5
Show file tree

Hide file tree

Showing 4 changed files with 18 additions and 2 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -9,10 +9,16 @@ Fixed:
 
   * segment-region: ensure polygons are within page/Border
 
+## [0.8.4] - 2020-06-05
+
 Changed:
 
   * segment-region: in `sparse_text` mode, also add text lines
 
+Fixed:
+
+  * Always set path to `TESSDATA_PREFIX` for `tesserocr.get_languages`, #129
+
 ## [0.8.3] - 2020-05-12
 
 Fixed:
@@ -182,6 +188,9 @@ Changed:
   * Recognition with proper support for textequiv_level, drop `page` level
 
 <!-- link-labels -->
+[0.8.4]: v0.8.4...v0.8.3
+[0.8.3]: v0.8.3...v0.8.2
+[0.8.2]: v0.8.2...v0.8.1
 [0.8.1]: v0.8.1...v0.8.0
 [0.8.0]: v0.8.0...v0.7.0
 [0.7.0]: v0.7.0...v0.6.0

diff --git a/Dockerfile b/Dockerfile
@@ -5,6 +5,7 @@ ENV PYTHONIOENCODING utf8
 
 WORKDIR /build-ocrd
 COPY setup.py .
+COPY ocrd_tesserocr/ocrd-tool.json .
 COPY README.md .
 COPY requirements.txt .
 COPY requirements_test.txt .

diff --git a/ocrd_tesserocr/ocrd-tool.json b/ocrd_tesserocr/ocrd-tool.json
@@ -1,5 +1,5 @@
 {
-  "version": "0.8.3",
+  "version": "0.8.4",
   "git_url": "https://github.com/OCR-D/ocrd_tesserocr",
   "dockerhub": "ocrd/tesserocr",
   "tools": {

diff --git a/ocrd_tesserocr/recognize.py b/ocrd_tesserocr/recognize.py
@@ -5,7 +5,7 @@
 
 from tesserocr import (
     RIL, PSM,
-    PyTessBaseAPI, get_languages)
+    PyTessBaseAPI, get_languages as get_languages_)
 
 from ocrd_utils import (
     getLogger,
@@ -46,6 +46,12 @@
 CHOICE_THRESHOLD_NUM = 6 # maximum number of choices to query and annotate
 CHOICE_THRESHOLD_CONF = 0.2 # maximum score drop from best choice to query and annotate
 
+def get_languages(*args, **kwargs):
+    """
+    Wraps tesserocr.get_languages() with a fixed path parameter.
+    """
+    return get_languages_(*args, path=TESSDATA_PREFIX, **kwargs)
+
 class TesserocrRecognize(Processor):
 
     def __init__(self, *args, **kwargs):