Add option to skip dictionary compile step (#201)

* Add option to skip dictionary compile step

Resolves #196

* Fix coverage

* Ensure we skip only if the dict doesn't exist

Author: facelessuser

.github/workflows/build.yml

@@ -90,11 +90,12 @@ jobs:
     # Results
     - name: Upload Results
       if: success()
-      uses: codecov/codecov-action@v3
+      uses: codecov/codecov-action@v4
       with:
         file: ./coverage.xml
         flags: unittests
         name: ${{ matrix.platform }}-${{ matrix.tox-env }}
+        token: ${{ secrets.CODECOV_TOKEN }} # required
         fail_ci_if_error: false
 
   docs_and_lint:
@@ -110,15 +111,15 @@ jobs:
     - name: Set up Python
       uses: actions/setup-python@v5
       with:
-        python-version: 3.x
+        python-version: '3.13'
     - name: Install Aspell
       if: matrix.tox-env == 'documents'
       run: |
         sudo apt-get install aspell aspell-en
     - name: Install dependencies
       run: |
         python -m pip install --upgrade pip
-        python -m pip install --upgrade build setuptools tox 
+        python -m pip install --upgrade build setuptools tox
     - name: ${{ matrix.tox-env }}
       run: |
         python -m tox -e ${{ matrix.tox-env }}

docs/src/markdown/about/changelog.md

@@ -1,5 +1,10 @@
 # Changelog
 
+## 2.11
+
+-   **NEW**: Add new command line option `--skip-dict-compile` which will skip the dictionary compiling step if the
+    dictionary already exists. Changes to a custom dictionary will be ignored.
+
 ## 2.10
 
 -   **NEW**: Allow specifying parallel processes to speed up spell checking. Number of jobs can be specified either

docs/src/markdown/index.md

@@ -53,26 +53,24 @@ If you want to manually install it, run `#!bash python setup.py build` and `#!ba
 ## Command Line Usage
 
 ```
-usage: pyspelling [-h] [--version] [--verbose] [--name NAME | --group GROUP] [--binary BINARY] [--jobs JOBS] [--config CONFIG] [--source SOURCE] [--spellchecker SPELLCHECKER]
+usage: pyspelling [-h] [--version] [--verbose] [--name NAME | --group GROUP] [--binary BINARY] [--jobs JOBS] [--config CONFIG] [--source SOURCE] [--spellchecker SPELLCHECKER] [--skip-dict-compile]
 
 Spell checking tool.
 
 options:
   -h, --help            show this help message and exit
   --version             show program's version number and exit
   --verbose, -v         Verbosity level.
-  --name NAME, -n NAME  Specific spelling task by name to run.
-  --group GROUP, -g GROUP
-                        Specific spelling task group to run.
-  --binary BINARY, -b BINARY
-                        Provide path to spell checker's binary.
-  --jobs JOBS, -j JOBS  Specify the number of spell checker processes to run in parallel.
-  --config CONFIG, -c CONFIG
-                        Spelling config.
-  --source SOURCE, -S SOURCE
-                        Specify override file pattern. Only applicable when specifying exactly one --name.
-  --spellchecker SPELLCHECKER, -s SPELLCHECKER
-                        Choose between aspell and hunspell
+  --name, -n NAME       Specific spelling task by name to run.
+  --group, -g GROUP     Specific spelling task group to run.
+  --binary, -b BINARY   Provide path to spell checker's binary.
+  --jobs, -j JOBS       Specify the number of spell checker processes to run in parallel.
+  --config, -c CONFIG   Spelling config.
+  --source, -S SOURCE   Specify override file pattern. Only applicable when specifying exactly one --name.
+  --spellchecker, -s SPELLCHECKER
+                        Choose between aspell and hunspell.
+  --skip-dict-compile, -x
+                        Skip dictionary compilation if the compiled file already exists.
 ```
 
 PySpelling can be run with the command below (assuming your Python bin/script folder is in your path).  By default it

pyspelling/__init__.py

@@ -578,7 +578,7 @@ class SpellingTask:
         "O": glob.O
     }
 
-    def __init__(self, checker, config, binary='', verbose=0, jobs=0, debug=False):
+    def __init__(self, checker, config, binary='', verbose=0, jobs=0, debug=False, skip_dict_compile=False):
         """Initialize."""
 
         if checker == "hunspell":  # pragma: no cover
@@ -594,6 +594,7 @@ def __init__(self, checker, config, binary='', verbose=0, jobs=0, debug=False):
         self.binary = checker if not binary else binary
         self.debug = debug
         self.jobs = jobs
+        self.skip_dict_compile = skip_dict_compile
 
     def log(self, text, level):
         """Log level."""
@@ -622,7 +623,13 @@ def walk_src(self, targets, flags, limit):
     def get_checker(self):
         """Get a spell checker object."""
 
-        checker = self.spellchecker(self.config, self.binary, self.verbose, self.default_encoding, self.debug)
+        checker = self.spellchecker(
+            self.config,
+            self.binary,
+            self.verbose,
+            self.default_encoding,
+            self.debug
+        )
         checker._build_pipeline(self.task)
         return checker
 
@@ -654,7 +661,15 @@ def run_task(self, task, source_patterns=None):
         self.task = task
         self.default_encoding = self.task.get('default_encoding', '')
         self.options = self.spellchecker.get_options(self.task)
-        self.personal_dict = self.spellchecker.setup_dictionary(self.task, self.binary, self.verbose)
+        if not self.skip_dict_compile:
+            self.personal_dict = self.spellchecker.setup_dictionary(self.task, self.binary, self.verbose)
+        else:
+            dictionary_options = self.task.get('dictionary', {})
+            output = os.path.abspath(dictionary_options.get('output', os.path.abspath(self.spellchecker.DICTIONARY)))
+            if os.path.exists(output):
+                self.personal_dict = output
+            else:
+                self.personal_dict = self.spellchecker.setup_dictionary(self.task, self.binary, self.verbose)
         self.found_match = False
         glob_flags = self._to_flags(self.task.get('glob_flags', "N|B|G"))
         glob_limit = self.task.get('glob_pattern_limit', 1000)
@@ -696,7 +711,8 @@ def spellcheck(
     sources=None,
     verbose=0,
     debug=False,
-    jobs=0
+    jobs=0,
+    skip_dict_compile=False
 ):
     """Spell check."""
 
@@ -737,7 +753,7 @@ def spellcheck(
 
         log('Using {} to spellcheck {}'.format(checker, task.get('name', '')), 1, verbose)
 
-        spelltask = SpellingTask(checker, config, binary, verbose, jobs, debug)
+        spelltask = SpellingTask(checker, config, binary, verbose, jobs, debug, skip_dict_compile)
 
         for result in spelltask.run_task(task, source_patterns=sources):
             log('Context: %s' % result.context, 2, verbose)

pyspelling/__main__.py

@@ -30,7 +30,13 @@ def main():
         help="Specify override file pattern. Only applicable when specifying exactly one --name."
     )
     parser.add_argument(
-        '--spellchecker', '-s', action='store', default='', help="Choose between aspell and hunspell"
+        '--spellchecker', '-s', action='store', default='', help="Choose between aspell and hunspell."
+    )
+    parser.add_argument(
+        '--skip-dict-compile',
+        '-x',
+        action='store_true',
+        help="Skip dictionary compilation if the compiled file already exists."
     )
     args = parser.parse_args()
 
@@ -44,6 +50,7 @@ def main():
         verbose=args.verbose,
         debug=args.debug,
         jobs=args.jobs,
+        skip_dict_compile=args.skip_dict_compile
     )
 
 
@@ -58,6 +65,7 @@ def run(config, **kwargs):
     sources = kwargs.get('sources', [])
     debug = kwargs.get('debug', False)
     jobs = kwargs.get('jobs', 0)
+    skip_dict_compile = kwargs.get('skip_dict_compile', False)
 
     fail = False
     count = 0
@@ -71,6 +79,7 @@ def run(config, **kwargs):
         verbose=verbose,
         debug=debug,
         jobs=jobs,
+        skip_dict_compile=skip_dict_compile
     ):
         count += 1
         if results.error:

tests/test_config.py

@@ -315,3 +315,70 @@ def test_too_many_filter_names(self):
         self.mktemp('test.txt', '', 'utf-8')
         with self.assertRaises(ValueError):
             self.assert_spellcheck('.source.yml', [])
+
+
+class TestSkipDictionary(util.PluginTestCase):
+    """Test no pipeline."""
+
+    def test_skip(self):
+        """Test text."""
+
+        config = self.dedent(
+            """
+            matrix:
+            - name: no_pipeline
+              default_encoding: utf-8
+              sources:
+              - '{temp}/**/*.txt'
+              aspell:
+                lang: en
+                d: en_US
+              hunspell:
+                d: en_US
+              pipeline: null
+              dictionary:
+                wordlists:
+                - '{temp}/mydict.wl'
+                output: '{temp}/mydict.dic'
+            """
+        ).format(temp=self.tempdir)
+        self.mktemp('.skip_compile.yml', config, 'utf-8')
+
+        bad_words = ['helo', 'begn']
+        good_words = ['yes', 'word']
+        self.mktemp('mydict.wl', '\n'.join(bad_words), 'utf-8')
+        self.mktemp('test.txt', '\n'.join(bad_words + good_words), 'utf-8')
+        self.assert_spellcheck('.skip_compile.yml', [], skip_dict_compile=True)
+
+    def test_compile_once(self):
+        """Test text."""
+
+        config = self.dedent(
+            """
+            matrix:
+            - name: no_pipeline
+              default_encoding: utf-8
+              sources:
+              - '{temp}/**/*.txt'
+              aspell:
+                lang: en
+                d: en_US
+              hunspell:
+                d: en_US
+              pipeline: null
+              dictionary:
+                wordlists:
+                - '{temp}/mydict.wl'
+                output: '{temp}/mydict.dic'
+            """
+        ).format(temp=self.tempdir)
+        self.mktemp('.skip_compile.yml', config, 'utf-8')
+
+        bad_words = ['helo', 'begn']
+        good_words = ['yes', 'word']
+        self.mktemp('mydict.wl', '\n'.join(bad_words), 'utf-8')
+        self.mktemp('test.txt', '\n'.join(bad_words + good_words), 'utf-8')
+        # For this to work, we need to run on either Hunspell or Aspell, not both as the dictionary
+        # will be overwritten with the format for the wrong spell checker.
+        self.assert_spellcheck('.skip_compile.yml', [], skip_dict_compile=False, only_one=True)
+        self.assert_spellcheck('.skip_compile.yml', [], skip_dict_compile=True, only_one=True)

tests/util.py

@@ -184,7 +184,17 @@ def assert_context(self, config_file, expected, names=None, groups=None, sources
                 context.append(results.context)
             self.assertEqual([(f'{self.tempdir}/{i}' if i else i) for i in expected[:]], context)
 
-    def assert_spellcheck(self, config_file, expected, names=None, groups=None, sources=None, verbose=4):
+    def assert_spellcheck(
+        self,
+        config_file,
+        expected,
+        names=None,
+        groups=None,
+        sources=None,
+        verbose=4,
+        skip_dict_compile=False,
+        only_one=False
+    ):
         """Spell check."""
 
         hunspell_location = which(HUNSPELL)
@@ -209,12 +219,15 @@ def assert_spellcheck(self, config_file, expected, names=None, groups=None, sour
                 checker='hunspell',
                 binary=hunspell_location,
                 debug=True,
-                verbose=verbose
+                verbose=verbose,
+                skip_dict_compile=skip_dict_compile
             ):
                 if results.error:
                     print(results.error)
                 words |= set(results.words)
             self.assertEqual(sorted(expected), sorted(words))
+            if only_one:
+                return
         if aspell_location:
             words = set()
             for results in spellcheck(
@@ -225,9 +238,12 @@ def assert_spellcheck(self, config_file, expected, names=None, groups=None, sour
                 checker='aspell',
                 binary=aspell_location,
                 debug=True,
-                verbose=verbose
+                verbose=verbose,
+                skip_dict_compile=skip_dict_compile
             ):
                 if results.error:
                     print(results.error)
                 words |= set(results.words)
             self.assertEqual(sorted(expected), sorted(words))
+            if only_one:
+                return