Moves to Python format strings for renaming, handles directory
structures, moving of files to a destination directory, sanitizes
file paths with pathvalidate and takes a different approach to
smart filename cleanup using the Python string.Formatter class
Moving to Python format strings means we can point to python
documentation for syntax and all we have to do is document the
properties and types that are attached to the GenericMetadata class.
Switching to pathvalidate allows comictagger to more simply handle both
directories and symbols in filenames.
The only changes to the string.Formatter class is:
1. format_field returns
an empty string if the value is none or an empty string regardless of
the format specifier.
2. _vformat drops the previous literal text if the field value
is an empty string and lstrips the following literal text of closing
special characters.
Fix file encoding inconsistencies, windows defaults to cp1252, which is
not Unicode compatible.
Add logging for all exceptions in the comicapi package
Ensure that all exceptions are logged and shown to the user
Remove no longer used google scripts
Remove convenience files from comicataggerlib and import comicapi directly
Add type-hints to facilitate auto-complete tools
Make PyQt5 code more compatible with PyQt6
Implement automatic tooling
isort and black for code formatting
Line length has been set to 120
flake8 for code standards with exceptions:
E203 - Whitespace before ':' - format compatiblity with black
E501 - Line too long - flake8 line limit cannot be set
E722 - Do not use bare except - fixing bare except statements is a
lot of overhead and there are already
many in the codebase
These changes, along with some manual fixes creates much more readable code.
See examples below:
diff --git a/comicapi/comet.py b/comicapi/comet.py
index d1741c5..52dc195 100644
--- a/comicapi/comet.py
+++ b/comicapi/comet.py
@@ -166,7 +166,2 @@ class CoMet:
- if credit['role'].lower() in set(self.editor_synonyms):
- ET.SubElement(
- root,
- 'editor').text = "{0}".format(
- credit['person'])
@@ -174,2 +169,4 @@ class CoMet:
self.indent(root)
+ if credit["role"].lower() in set(self.editor_synonyms):
+ ET.SubElement(root, "editor").text = str(credit["person"])
diff --git a/comictaggerlib/autotagmatchwindow.py b/comictaggerlib/autotagmatchwindow.py
index 4338176..9219f01 100644
--- a/comictaggerlib/autotagmatchwindow.py
+++ b/comictaggerlib/autotagmatchwindow.py
@@ -63,4 +63,3 @@ class AutoTagMatchWindow(QtWidgets.QDialog):
self.skipButton, QtWidgets.QDialogButtonBox.ActionRole)
- self.buttonBox.button(QtWidgets.QDialogButtonBox.Ok).setText(
- "Accept and Write Tags")
+ self.buttonBox.button(QtWidgets.QDialogButtonBox.StandardButton.Ok).setText("Accept and Write Tags")
diff --git a/comictaggerlib/cli.py b/comictaggerlib/cli.py
index 688907d..dbd0c2e 100644
--- a/comictaggerlib/cli.py
+++ b/comictaggerlib/cli.py
@@ -293,7 +293,3 @@ def process_file_cli(filename, opts, settings, match_results):
if opts.raw:
- print((
- "{0}".format(
- str(
- ca.readRawCIX(),
- errors='ignore'))))
+ print(ca.read_raw_cix())
else:
Because additional series results are now returned due to #143 the series selection window can with a large number of results that are not usually sorted in a useful way.
I've created 3 settings that can help finding the corect series quickly
use the publisher black list - can be toggled from the series selction screen, as well as a setting for is default behaviour
a setting to make the result initially sorted by start year instead of the default no of issues
a setting to initially put exact and near matches at the top of the list
Refactor removearticles to only remove articles
Add normalization on the search string and the series name results
Searching now only compares ASCII a-z and 0-9 and all other characters
are replaced with single space, this is done to both the search string
and the result. This fixes an with names that are separated by a
hyphen (-) in the filename but in the Comic Vine name are separated by a
slash (/) and other similar issues.
