bhengh
Member
|
I would like to use Programmer’s Notepad to write Python apps. Is there anyway to get Programmer’s Notepad to auto-indent after a block statement?
|
simon
Key Master
|
Indeed, and PyPN already comes with auto-indent support for Python, so you don’t need to write your own indenter.
|
CoDEmanX
Member
|
Here is the code:
PN_dirini.py
@indenter("python")
def python_indent(c, doc):
sci = scintilla.Scintilla(doc)
if c == 'n' or c == 'r':
pos = sci.CurrentPos
line = sci.LineFromPosition( pos )
lc = findPrevLineLastChar( pos, sci )
# If the previous line ended with a colon, then indent
if lc == ':':
indent = sci.GetLineIndentation( line )
# The DumbIndent system may already have indented this line...
previndent = sci.GetLineIndentation( line - 1 )
if indent == previndent or indent == 0:
indent += 4
sci.IndentLine( line, indent )
Next stable release should contain indenters for the most used languages out of the box, that’d be awesome!
|
CoDEmanX
Member
|
Just came to my mind:
Python indent won’t work if you enter a comment after a colon!
Example:
if condition < 1: # check something
no auto-indent, as previous line's last character isn't a colon
On the other hand, it may indent for no reason, if a comment ends with a colon
call_function() # smiley (:
stupid auto-indent
|
simon
Key Master
|
Yes, there are a couple of bugs in the python indenter. Comments is one, and the other is it will keep indenting if you add extra lines below the indent point. i.e. this:
type: def monkey():
type: [enter]
type: [enter]
Would be happy to accept patches for any of these issues or other indenters!
|
CoDEmanX
Member
|
Lol, it bet it indents more and more ’cause it ignores whitespace in the findPrevLineLastChar function:
def findPrevLineLastChar(p, sci):
while p > 0:
p = p - 1
c = chr( sci.GetCharAt(p) )
# Look for a non-whitespace character ending the previous line.
if not c in ['n','r','t',' ']:
return c
return None
But simply removing the white-space check won’t work, nor dropping just the ‘t’. It should check whether the whole previous line is white-space or comment and go back further until it fetches the right character.
Or perhaps do a completely new algorithm. I made an indentation script for C++, which went all over the file and increased indentation on the next line after { (it kept track of the opening and closing braces). Dunno if this would work for Python, as there are no closing brackets or end-keyword…
|
fongandrew
Member
|
In case anyone is stumbling on this via Google, here’s a fix for the comment and multiple indentation bugs. Edit PN_dirinit.py and change the python_indent function as follows:
import re
# Indention regex - look for colon, then optional spaces and comment
PY_INDENT_RE = re.compile(r':s*(#.*)?$')
@indenter("python")
def python_indent(c, doc):
sci = scintilla.Scintilla(doc)
if c == 'n' or c == 'r':
pos = sci.CurrentPos
line = sci.LineFromPosition( pos )
prev_line = sci.GetLine(line - 1)
# If the previous line matches regular expression above, then indent
if PY_INDENT_RE.search(prev_line):
indent = sci.GetLineIndentation( line )
# The DumbIndent system may already have indented this line...
previndent = sci.GetLineIndentation( line - 1 )
if indent == previndent or indent == 0:
indent += 4
sci.IndentLine( line, indent )
|
CoDEmanX
Member
|
i just tried your code but it doesn’t seem to change anything, still indenting everything without exception if previous line ends with a colon.
Do i have to delete a cached version of the init.py script somehow? But where would it be located at?
|
simon
Key Master
|
If you have PN installed (rather than portable) then you might need to edit the script using PN running as administrator. Otherwise Windows Vista/7 will redirect your save to a “special” folder where PN won’t see it when compiling scripts.
|
CoDEmanX
Member
|
Did that already, i guess the regex just doesn’t work as it should. With PN regex search, it still matches colons in comments if it’s the last character or if it’s followed by white-space only.
|
NickDMax
Member
|
try: PY_INDENT_RE = re.compile(r'^[^#:]*:[^"']*$')
the basic logic is: find a colon that is not in a comment, and then ensure it is not in a string or doc comment.
|
CoDEmanX
Member
|
great! it’s doing indentation for code-colons and nothing for comment-colons even if it’s at the very end
however, it indents code-colons in “”" comments “”", but that’s not much of a problem as i’m mostly using single line #comments!
"""
for x in y:
indented!
exec() # foo: bar
not indented!
"""
|
simon
Key Master
|
Unfortunately python’s regexs aren’t guaranteed to match the ones PN uses. Test them with python itself to be sure of the one you’re using.
|