PathFilter questions

[q3cpma]

Hello,

as a POSIX OS user, I'm wondering about the forced character stripping in paths, and am confused at the PathFilter class in path.py:

• What are these "separators" (: and |) and why are they stripped?
• Same for "special" characters, since fat is already here.

I'm thinking about replacing everything but quotes by "POSIX" and "vfat", while having a configuration option for adding an additional string of characters to replace and maybe another for the fallback character.

[MerlijnWajer]

This looks weird indeed. The unicode quoting code is also something we can get rid of, I think.

[q3cpma]

On Tue, Oct 23, 2018 at 03:11:10AM -0700, Merlijn Wajer wrote: This looks weird indeed. The unicode quoting code is also something we can get rid of, I think. -- You are receiving this because you authored the thread. Reply to this email directly or view it on GitHub: #313 (comment)

What about something like this? This would leave only the configuration question (autodetect during the first run and store in the config file?) diff --git a/whipper/common/path.py b/whipper/common/path.py index a5d0eb4..2b0fba9 100644 --- a/whipper/common/path.py +++ b/whipper/common/path.py @@ -21,47 +21,24 @@ import re +POSIX = 0 +MACOS = 1 +WINDOWS = 2 + class PathFilter(object): """ I filter path components for safe storage on file systems. """ - def __init__(self, slashes=True, quotes=True, fat=True, special=False): + def __init__(self, os=POSIX): """ - @param slashes: whether to convert slashes to dashes - @param quotes: whether to normalize quotes - @param fat: whether to strip characters illegal on FAT filesystems - @param special: whether to strip special characters + @param os: the path character set restrictions to use """ - self._slashes = slashes - self._quotes = quotes - self._fat = fat - self._special = special + self._os = os def filter(self, path): - if self._slashes: - path = re.sub(r'[/\\]', '-', path, re.UNICODE) - - def separators(path): - # replace separators with a space-hyphen or hyphen - path = re.sub(r'[:]', ' -', path, re.UNICODE) - path = re.sub(r'[\|]', '-', path, re.UNICODE) - return path - - # change all fancy single/double quotes to normal quotes - if self._quotes: - path = re.sub(ur'[\xc2\xb4\u2018\u2019\u201b]', "'", path, - re.UNICODE) - path = re.sub(ur'[\u201c\u201d\u201f]', '"', path, re.UNICODE) - - if self._special: - path = separators(path) - path = re.sub(r'[\*\?&!\'\"\$\(\)`{}\[\]<>]', - '_', path, re.UNICODE) - - if self._fat: - path = separators(path) - # : and | already gone, but leave them here for reference - path = re.sub(r'[:\*\?"<>|"]', '_', path, re.UNICODE) - - return path + if self._os == POSIX or self._os == MACOS: + repl = r'/' + else: + repl = r'[/\\:*?"<>|]' + return re.sub(repl, '_', path, re.UNICODE)

[MerlijnWajer]

I think you can use sys.platform for that

[JoeLametta]

@q3cpma Let me quote you to have the diff formatted in a code block:

diff --git a/whipper/common/path.py b/whipper/common/path.py
index a5d0eb4..2b0fba9 100644
--- a/whipper/common/path.py
+++ b/whipper/common/path.py
@@ -21,47 +21,24 @@
 import re


+POSIX = 0
+MACOS = 1
+WINDOWS = 2
+
 class PathFilter(object):
     """
     I filter path components for safe storage on file systems.
     """

-    def __init__(self, slashes=True, quotes=True, fat=True, special=False):
+    def __init__(self, os=POSIX):
         """
-        @param slashes: whether to convert slashes to dashes
-        @param quotes:  whether to normalize quotes
-        @param fat:     whether to strip characters illegal on FAT filesystems
-        @param special: whether to strip special characters
+        @param os: the path character set restrictions to use
         """
-        self._slashes = slashes
-        self._quotes = quotes
-        self._fat = fat
-        self._special = special
+        self._os = os

     def filter(self, path):
-        if self._slashes:
-            path = re.sub(r'[/\\]', '-', path, re.UNICODE)
-
-        def separators(path):
-            # replace separators with a space-hyphen or hyphen
-            path = re.sub(r'[:]', ' -', path, re.UNICODE)
-            path = re.sub(r'[\|]', '-', path, re.UNICODE)
-            return path
-
-        # change all fancy single/double quotes to normal quotes
-        if self._quotes:
-            path = re.sub(ur'[\xc2\xb4\u2018\u2019\u201b]', "'", path,
-                          re.UNICODE)
-            path = re.sub(ur'[\u201c\u201d\u201f]', '"', path, re.UNICODE)
-
-        if self._special:
-            path = separators(path)
-            path = re.sub(r'[\*\?&!\'\"\$\(\)`{}\[\]<>]',
-                          '_', path, re.UNICODE)
-
-        if self._fat:
-            path = separators(path)
-            # : and | already gone, but leave them here for reference
-            path = re.sub(r'[:\*\?"<>|"]', '_', path, re.UNICODE)
-
-        return path
+        if self._os == POSIX or self._os == MACOS:
+            repl = r'/'
+        else:
+            repl = r'[/\\:*?"<>|]'
+        return re.sub(repl, '_', path, re.UNICODE)