Merge remote-tracking branch 'upstream/develop' into detect-fasttext-…

…in-PATH

.github/workflows/tests.yml

@@ -16,11 +16,9 @@ jobs:
       fail-fast: false
       matrix:
         include:
-          - {name: Linux, python: 3.6, os: ubuntu-20.04, tox: 'flake8,flake8-docs'}
-          - {name: Linux, python: 3.6, os: ubuntu-20.04, tox: 'py36-linux'}
+          - {name: Linux, python: 3.7, os: ubuntu-20.04, tox: 'flake8,flake8-docs'}
           - {name: Linux, python: 3.7, os: ubuntu-20.04, tox: 'py37-linux'}
-          - {name: Linux, python: 3.8, os: ubuntu-20.04, tox: 'py38-linux'}
-          - {name: Windows, python: 3.6, os: windows-2019, tox: 'py36-win'}
+          - {name: Linux, python: 3.8, os: ubuntu-20.04, tox: 'py38-linux-cov'}
           - {name: Windows, python: 3.7, os: windows-2019, tox: 'py37-win'}
           - {name: Windows, python: 3.8, os: windows-2019, tox: 'py38-win'}
     env:
@@ -72,4 +70,4 @@ jobs:
         run: |
           pwd
           COREFILE=$(find . -maxdepth 1 -name "core*" | head -n 1)
-          if [[ -f "$COREFILE" ]]; then EXECFILE=$(gdb -c "$COREFILE" -batch | grep "Core was generated" | tr -d "\`" | cut -d' ' -f5); file "$COREFILE"; gdb -c "$COREFILE" "$EXECFILE" -x continuous_integration/debug.gdb -batch; fi
+          if [[ -f "$COREFILE" ]]; then EXECFILE=$(gdb -c "$COREFILE" -batch | grep "Core was generated" | tr -d "\`" | cut -d' ' -f5); file "$COREFILE"; gdb -c "$COREFILE" "$EXECFILE" -x continuous_integration/debug.gdb -batch; fi

CHANGELOG.md

@@ -10,6 +10,7 @@ Changes
 * [#3250](https://github.com/RaRe-Technologies/gensim/pull/3250): Make negative ns_exponent work correctly, by [@menshikh-iv](https://github.com/menshikh-iv)
 * [#3258](https://github.com/RaRe-Technologies/gensim/pull/3258): Adding another check to _check_corpus_sanity for compressed files, adding test, by [@dchaplinsky](https://github.com/dchaplinsky)
 * [#3274](https://github.com/RaRe-Technologies/gensim/pull/3274): Migrate setup.py from distutils to setuptools, by [@geojacobm6](https://github.com/geojacobm6)
+* [#3286](https://github.com/RaRe-Technologies/gensim/pull/3286): Fixes 'not enough arguments for format string' error, by [@gilbertfrancois](https://github.com/gilbertfrancois)
 
 ## 4.1.2, 2021-09-17
 

gensim/models/doc2vec.py

@@ -993,7 +993,9 @@ def _scan_vocab(self, corpus_iterable, progress_per, trim_rule):
             logger.warning(
                 "Highest int doctag (%i) larger than count of documents (%i). This means "
                 "at least %i excess, unused slots (%i bytes) will be allocated for vectors.",
-                max_rawint, corpus_count, ((max_rawint - corpus_count) * self.vector_size * 4))
+                max_rawint, corpus_count, max_rawint - corpus_count,
+                (max_rawint - corpus_count) * self.vector_size * dtype(REAL).itemsize,
+            )
         if max_rawint > -1:
             # adjust indexes/list to account for range of pure-int keyed doctags
             for key in doctags_list:

tox.ini

@@ -1,6 +1,6 @@
 [tox]
 minversion = 2.0
-envlist = {py36,py37,py38, py39}-{win,linux}, flake8, docs, docs-upload, download-wheels, upload-wheels, test-pypi
+envlist = {py36,py37,py38, py39}-{win,linux}, py38-linux-cov, flake8, docs, docs-upload, download-wheels, upload-wheels, test-pypi
 skipsdist = True
 platform = linux: linux
            win: win64
@@ -42,9 +42,11 @@ exclude_lines =
 
 ignore_errors = True
 
+#
+# Conditional factors https://tox.wiki/en/latest/config.html#factors
+#
 [pytest]
-addopts = -rfxEXs --durations=20 --showlocals --cov=gensim/ --cov-report=xml
-
+addopts = -rfxEXs --durations=20 --showlocals
 
 [testenv]
 recreate = True
@@ -72,7 +74,8 @@ commands =
     python --version
     pip --version
     python setup.py build_ext --inplace
-    pytest {posargs:gensim/test}
+    cov: pytest {posargs:gensim/test} --cov=gensim/ --cov-report=xml
+    !cov: pytest {posargs:gensim/test}
 
 
 [testenv:flake8]