Merge pull request #125 from dynobo/dev

Dev

.github/workflows/test.yaml

@@ -66,7 +66,9 @@ jobs:
       - name: Pytest on Linux
         if: ${{ matrix.os  == 'ubuntu-20.04' }}
         # xvfb-run with bpp 32 needed to grab screenshot with mss
-        run: xvfb-run --server-args="-screen 0 1024x768x24+32" poetry run pytest
+        run: |
+          xvfb-run --server-args="-screen 0 1024x768x24+32" \
+            poetry run pytest -m "not skip_on_gh_linux"
         env:
           XDG_SESSION_TYPE: wayland
       - name: Pytest on Windows

CHANGELOG.md

@@ -1,5 +1,11 @@
 # Changelog
 
+## v0.2.3 (2021-07-16)
+
+**Changes:**
+
+- Fix styling issues with settings menu
+
 ## v0.2.2 (2021-07-15)
 
 **Changes:**

FAQ.md

@@ -1,4 +1,4 @@
-# Frequently Asked Questions  <!-- omit in toc -->
+# Frequently Asked Questions <!-- omit in toc -->
 
 - [General](#general)
   - [How can I add additional languages to NormCap?](#how-can-i-add-additional-languages-to-normcap)
@@ -31,10 +31,15 @@ to install additional language for Tesseract on your system.
 
 ### What can I do improve the detection accuracy?
 
-The most import thing is to
-[specify the correct language](#how-can-i-specify-the-languages-to-be-detected). If you
-the results are still quite bad, please submit a screenshot of the text your are trying
-to recognize [as an issue](https://github.com/dynobo/normcap/issues).
+The most import thing is to specify the correct language (via settings menu or the
+`--language` command line argument). But keep in mind, that selecting multiple languages
+at once slows down the recognition a bit.
+
+Tesseract sometimes also struggles with recognizing text with just very view characters,
+like a single word. In this case it might help to select a larger portion of text.
+
+If you the results are still quite bad, please submit a screenshot of the text your are
+trying to recognize [as an issue](https://github.com/dynobo/normcap/issues).
 
 ### Is my image analyzed in "the cloud"?
 

README.md

@@ -118,7 +118,10 @@ normcap
 
 ### Command line options
 
-NormCap has no settings, just a set of command line arguments:
+NormCap has some settings. If you use the settings menu in the UI, your settings are
+persisted in `<users config folder>/normcap/config.yaml`. If you start NormCap with a
+command line argument (other then `-h`, `-v` and `-V`), all persisted settings are
+ignored and only the settings provided as arguments are used:
 
 ```plain
 (normcap)dynobo@cioran:~$ normcap --help

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "normcap"
-version = "0.2.2"
+version = "0.2.3"
 description = "OCR-powered screen-capture tool to capture information instead of images."
 keywords = ["screenshot", "ocr", "capture", "clipboard"]
 license = "GPLv3"
@@ -94,6 +94,7 @@ disable = [
     "invalid-name",
     "logging-fstring-interpolation",
     "logging-not-lazy",
+    "missing-module-docstring",
     "too-few-public-arguments",
     "too-few-public-methods",
     "too-many-arguments",
@@ -112,7 +113,7 @@ build-backend = "poetry.core.masonry.api"
 [tool.briefcase]
 project_name = "NormCap"
 bundle = "eu.dynobo"
-version = "0.2.2"
+version = "0.2.3"
 url = "https://github.com/dynobo/normcap"
 license = "GPLv3"
 author = 'dynobo'

src/normcap/clipboard.py

@@ -10,7 +10,7 @@ def init():
 
     app = QtWidgets.QApplication.instance()
     if app is None:
-        app = QtWidgets.QApplication([])
+        app = QtWidgets.QApplication()
 
     def copy_qt(text):
         cb = app.clipboard()