]> gitweb @ CieloNegro.org - youtube-dl.git/commitdiff
[README.md] Move code from #21939 to the right place
authorSergey M <dstftw@gmail.com>
Thu, 1 Aug 2019 22:54:56 +0000 (05:54 +0700)
committerGitHub <noreply@github.com>
Thu, 1 Aug 2019 22:54:56 +0000 (05:54 +0700)
README.md

index 8c48a30121bdd53344868615bced9ce00b99e40b..c39b13616946fbb1c07377e1bc74a39a3dfc4ee5 100644 (file)
--- a/README.md
+++ b/README.md
@@ -1216,6 +1216,72 @@ Incorrect:
 'PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4'
 ```
 
+### Inline values
+
+Extracting variables is acceptable for reducing code duplication and improving readability of complex expressions. However, you should avoid extracting variables used only once and moving them to opposite parts of the extractor file, which makes reading the linear flow difficult.
+
+#### Example
+
+Correct:
+
+```python
+title = self._html_search_regex(r'<title>([^<]+)</title>', webpage, 'title')
+```
+
+Incorrect:
+
+```python
+TITLE_RE = r'<title>([^<]+)</title>'
+# ...some lines of code...
+title = self._html_search_regex(TITLE_RE, webpage, 'title')
+```
+
+### Collapse fallbacks
+
+Multiple fallback values can quickly become unwieldy. Collapse multiple fallback values into a single expression via a list of patterns.
+
+#### Example
+
+Good:
+
+```python
+description = self._html_search_meta(
+    ['og:description', 'description', 'twitter:description'],
+    webpage, 'description', default=None)
+```
+
+Unwieldy:
+
+```python
+description = (
+    self._og_search_description(webpage, default=None)
+    or self._html_search_meta('description', webpage, default=None)
+    or self._html_search_meta('twitter:description', webpage, default=None))
+```
+
+Methods supporting list of patterns are: `_search_regex`, `_html_search_regex`, `_og_search_property`, `_html_search_meta`.
+
+### Trailing parentheses
+
+Always move trailing parentheses after the last argument.
+
+#### Example
+
+Correct:
+
+```python
+    lambda x: x['ResultSet']['Result'][0]['VideoUrlSet']['VideoUrl'],
+    list)
+```
+
+Incorrect:
+
+```python
+    lambda x: x['ResultSet']['Result'][0]['VideoUrlSet']['VideoUrl'],
+    list,
+)
+```
+
 ### Use convenience conversion and parsing functions
 
 Wrap all extracted numeric data into safe functions from [`youtube_dl/utils.py`](https://github.com/ytdl-org/youtube-dl/blob/master/youtube_dl/utils.py): `int_or_none`, `float_or_none`. Use them for string to number conversions as well.