[gedit-plugins] Corrected align whitespace groups
- From: Jesse van den Kieboom <jessevdk src gnome org>
- To: commits-list gnome org
- Cc:
- Subject: [gedit-plugins] Corrected align whitespace groups
- Date: Thu, 16 Aug 2012 08:12:37 +0000 (UTC)
commit ce58f1dc6a449fc89273398c1cb5cec5aeefe008
Author: Jesse van den Kieboom <jesse vandenkieboom epfl ch>
Date: Thu Aug 16 10:12:27 2012 +0200
Corrected align whitespace groups
plugins/commander/modules/align.py | 173 +++++++++++++++++++++---------------
1 files changed, 100 insertions(+), 73 deletions(-)
---
diff --git a/plugins/commander/modules/align.py b/plugins/commander/modules/align.py
index ac4a219..68c1a76 100644
--- a/plugins/commander/modules/align.py
+++ b/plugins/commander/modules/align.py
@@ -28,15 +28,38 @@ import re
__commander_module__ = True
+def _get_groups(m, group, add_ws_group):
+ if len(m.groups()) <= group - 1:
+ gidx = 0
+ else:
+ gidx = group
+
+ if len(m.groups()) <= add_ws_group - 1:
+ wsidx = 0
+ else:
+ wsidx = add_ws_group
+
+ # Whitespace group must be contained in align group
+ if m.start(wsidx) < m.start(gidx) or m.end(wsidx) > m.end(gidx):
+ wsidx = gidx
+
+ return (gidx, wsidx)
+
class Line:
def __init__(self, line, reg, tabwidth):
self.tabwidth = tabwidth
self.line = line
+
+ # All the separators
self.matches = list(reg.finditer(line))
+ # @newline initially contains the first column
+
if not self.matches:
+ # No separator found
self.newline = str(line)
else:
+ # Up to first separator
self.newline = line[0:self.matches[0].start(0)]
def matches_len(self):
@@ -57,33 +80,23 @@ class Line:
if m == None:
return
- if len(m.groups()) <= group - 1:
- gidx = 0
- else:
- gidx = group
-
- if len(m.groups()) <= add_ws_group - 1:
- wsidx = 0
- else:
- wsidx = add_ws_group
+ gidx, wsidx = _get_groups(m, group, add_ws_group)
- if wsidx > gidx:
- wsidx = gidx
+ # Append leading match
+ self.newline += self.line[m.start(0):m.start(gidx)]
- # First append leading match
- wsg = m.group(wsidx)
- numln = len(wsg) - len(wsg.lstrip())
+ # Now align by replacing wsidx with spaces
+ prefix = self.line[m.start(gidx):m.start(wsidx)]
+ suffix = self.line[m.end(wsidx):m.end(gidx)]
+ sp = ''
- self.newline += self.line[m.start(0):m.start(wsidx)]
+ while True:
+ bridge = prefix + sp + suffix
- # Then do the indent until gidx is aligned with num
- bridge = self.line[m.start(wsidx) + numln:m.start(gidx)]
-
- if not '\t' in bridge:
- bridge = (' ' * (num - self.new_len(bridge))) + bridge
- else:
- while self.new_len(bridge) < num:
- bridge = ' ' + bridge
+ if self.new_len(bridge) < num:
+ sp += ' '
+ else:
+ break
self.newline += bridge
@@ -95,46 +108,23 @@ class Line:
else:
endidx = mnext.start(0)
- self.newline += self.line[m.start(gidx):endidx]
+ self.newline += self.line[m.end(gidx):endidx]
def __str__(self):
return self.newline
-def __default__(view):
- """Align selected text in columns: align
-
-Align the selected text in columns separated by white space (spaces and tabs)"""
- yield regex(view, '\s+')
-
def _find_max_align(lines, idx, group, add_ws_group):
num = 0
- spaces = re.compile('^\s*')
# We will align on 'group', by adding spaces to 'add_ws_group'
for line in lines:
m = line.match(idx)
if m != None:
- if len(m.groups()) <= group - 1:
- gidx = 0
- else:
- gidx = group
-
- if len(m.groups()) <= add_ws_group - 1:
- wsidx = 0
- else:
- wsidx = add_ws_group
-
- if wsidx > gidx:
- wsidx = gidx
-
- wsg = m.group(wsidx)
-
- numln = len(wsg) - len(wsg.lstrip())
+ gidx, wsidx = _get_groups(m, group, add_ws_group)
# until the start
- extra = line.line[m.start(0):m.start(wsidx)]
- extra += line.line[m.start(wsidx) + numln:m.start(gidx)]
+ extra = line.line[m.start(0):m.start(wsidx)] + line.line[m.end(wsidx):m.end(gidx)]
# Measure where to align it
l = line.new_len(extra)
@@ -149,19 +139,37 @@ def _find_max_align(lines, idx, group, add_ws_group):
def _regex(view, reg, group, additional_ws, add_ws_group, flags=0):
buf = view.get_buffer()
+ # Get the selection of lines to align columns on
bounds = buf.get_selection_bounds()
- if not bounds or (bounds[1].get_line() - bounds[0].get_line() == 0):
- raise commander.commands.exceptions.Execute('You need to select a number of lines to align')
+ if not bounds:
+ start = buf.get_iter_at_mark(buf.get_insert())
+ start.set_line_offset(0)
+
+ end = start.copy()
+
+ if not end.ends_line():
+ end.forward_to_line_end()
+
+ bounds = (start, end)
+
+ if not bounds[0].equal(bounds[1]) and bounds[1].starts_line():
+ bounds[1].backward_line()
+ if not bounds[1].ends_line():
+ bounds[1].forward_to_line_end()
+
+ # Get the regular expression from the user
if reg == None:
reg, words, modifier = (yield commander.commands.result.Prompt('Regex:'))
+ # Compile the regular expression
try:
reg = re.compile(reg, flags)
except Exception, e:
raise commander.commands.exceptions.Execute('Failed to compile regular expression: %s' % (e,))
+ # Query the user to provide a regex group number to align on
if group == None:
group, words, modifier = (yield commander.commands.result.Prompt('Group (1):'))
@@ -170,22 +178,27 @@ def _regex(view, reg, group, additional_ws, add_ws_group, flags=0):
except:
group = 1
+ # Query the user for additional whitespace to insert for separating items
if additional_ws == None:
- additional_ws, words, modifier = (yield commander.commands.result.Prompt('Additional Whitespace (0):'))
+ additional_ws, words, modifier = (yield commander.commands.result.Prompt('Additional whitespace (0):'))
try:
additional_ws = int(additional_ws)
except:
additional_ws = 0
+ # Query the user for the regex group number on which to add the
+ # whitespace
if add_ws_group == None:
- add_ws_group, words, modifier = (yield commander.commands.result.Prompt('Group (1):'))
+ add_ws_group, words, modifier = (yield commander.commands.result.Prompt('Whitespace group (1):'))
try:
add_ws_group = int(add_ws_group)
except:
add_ws_group = -1
+ # By default, add the whitespace on the group on which the columns are
+ # aligned
if add_ws_group < 0:
add_ws_group = group
@@ -202,13 +215,11 @@ def _regex(view, reg, group, additional_ws, add_ws_group, flags=0):
num = 0
tabwidth = view.get_tab_width()
- for line in lines:
- ln = Line(line, reg, tabwidth)
-
- newlines.append(ln)
+ # Construct Line objects for all the lines
+ newlines = [Line(line, reg, tabwidth) for line in lines]
- if len(ln.matches) > num:
- num = len(ln.matches)
+ # Calculate maximum number of matches (i.e. columns)
+ num = reduce(lambda x, y: max(x, y.matches_len()), newlines, 0)
for i in range(num):
al = _find_max_align(newlines, i, group, add_ws_group)
@@ -221,35 +232,51 @@ def _regex(view, reg, group, additional_ws, add_ws_group, flags=0):
buf.begin_user_action()
buf.delete(bounds[0], bounds[1])
+
+ m = buf.create_mark(None, bounds[0], True)
+
buf.insert(bounds[1], aligned)
+ buf.select_range(buf.get_iter_at_mark(m), bounds[1])
+ buf.delete_mark(m)
+
buf.end_user_action()
yield commander.commands.result.DONE
-def regex(view, reg=None, group=1, additional_ws=1, add_ws_group=-1):
- """Align selected in columns using a regular expression: align.regex [<regex>] [<group>] [<ws>]
+def __default__(view, reg='\s+', align_group=1, padding=1, padding_group=-1):
+ """Align selected in columns using a regular expression: align.regex [<regex>=<strong>\s+</strong>] [<align-group>] [<padding>] [<padding-group>=<strong><align-group></strong>]
Align the selected text in columns separated by the specified regular expression.
-The optional group argument specifies on which group in the regular expression
-the text should be aligned. Additional whitespace will be added in front of
-the matched group to align each column. The optional <ws> argument can
-be used to add additional whitespace to the column separation.
+
+The optional <align-group> argument specifies on which group in the regular expression
+the text should be aligned and defaults to 1 (or 0 in the case that there is
+no explicit group specified). The <align-group> will be <em>replaced</em>
+with whitespace to align the columns. The optional <padding> argument can
+be used to add additional whitespace to the column separation. The last
+optional argument (<padding-group>) can be used to specify a separate
+group (which must be contained in <align-group>) which to replace with
+whitespace.
The regular expression will be matched in case-sensitive mode"""
- yield _regex(view, reg, group, additional_ws, add_ws_group)
+ yield _regex(view, reg, align_group, padding, padding_group)
-def regex_i(view, reg=None, group=1, additional_ws=1, add_ws_group=-1):
- """Align selected in columns using a regular expression: align.regex [<regex>] [<group>] [<ws>]
+def i(view, reg='\s+', align_group=1, padding=1, padding_group=-1):
+ """Align selected in columns using a regular expression: align.regex [<regex>=<strong>\s+</strong>] [<align-group>] [<padding>] [<padding-group>=<strong><align-group></strong>]
Align the selected text in columns separated by the specified regular expression.
-The optional group argument specifies on which group in the regular expression
-the text should be aligned. Additional whitespace will be added in front of
-the matched group to align each column. The optional <ws> argument can
-be used to add additional whitespace to the column separation.
+
+The optional <align-group> argument specifies on which group in the regular expression
+the text should be aligned and defaults to 1 (or 0 in the case that there is
+no explicit group specified). The <align-group> will be <em>replaced</em>
+with whitespace to align the columns. The optional <padding> argument can
+be used to add additional whitespace to the column separation. The last
+optional argument (<padding-group>) can be used to specify a separate
+group (which must be contained in <align-group>) which to replace with
+whitespace.
The regular expression will be matched in case-insensitive mode"""
- yield _regex(view, reg, group, additional_ws, add_ws_group, re.IGNORECASE)
+ yield _regex(view, reg, align_group, padding, padding_group, re.IGNORECASE)
# vi:ex:ts=4:et
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]
