(* Adobe's PageMill 1.0 is a nearly-great application for quickly and efficiently
creating World Wide Web pages which look good in common browsers such as Netscape.
It is only *nearly*-great, however, since it makes its own set of assumptions
and some down-right mistakes in HTML code. The following script will open PageMill
documents using BBEdit 3.5 (the commercial version only -- I don't think the Lite
version supports AppleScript).
Some of the Find/Replace patterns change true errors in the PageMill code
(significantly, Pattern (10) changes the notorious
code to the more
standard ). However, many of the Find/Replace patterns are cosmetic only,
and do not affect the final HTML code.
Important!!!!
Note that if you, for any reason want to keep runs of multiple spaces (say to
preserve whitespace using the
tag), you must comment out the
fourth pattern!
The following script is "e-mail-ware." If you like it, send me a friendly message.
if you have suggestions for improvements, I want to hear from you even more!
Clint MacDonald
Cell Biology & Biochemistry
Texas Tech University Health Sciences Center
3601 4th Street
Lubbock, TX 79430
March 27, 1996 *)
-- ----------
property HTML_File : false
property HTML_File_End : false
global docList
-- ----------
on run
tell application "Finder"
activate
set docList to selection
end tell
tell application "BBEdit 3.5"
activate
repeat with i from 1 to (count items in docList)
open (item i of docList as alias)
checkHTML() of me
if HTML_File and HTML_File_End then
cleanUpPageMill(docList) of me
else
display dialog "The file " & (item i of docList) & Â
" does not seem to be an HTML file." buttons {"OK"} default button 1
tell me to quit
end if
end repeat
display dialog Â
"Would you like to save the changes to all files?" buttons {"Yes", "Later"} default button 2
if button returned of result is "Yes" then
repeat with i from 1 to (count of windows)
go to line 0
tell window i to save
end repeat
else
go to line 0
end if
end tell -- BBEdit
end run
-- ----------
on open (docList)
tell application "BBEdit 3.5"
activate
repeat with i from 1 to (count items in docList)
open (item i of docList as alias)
checkHTML() of me
if HTML_File and HTML_File_End then
cleanUpPageMill(docList) of me
else
display dialog "The file " & (item i of docList) & Â
" does not seem to be an HTML file." buttons {"OK"} default button 1
tell me to quit
end if
end repeat
display dialog Â
"Would you like to save the changes to all files?" buttons {"Yes", "Later"} default button 2
if button returned of result is "Yes" then
repeat with i from 1 to (count of windows)
go to line 0
tell window i to save
end repeat
else
go to line 0
end if
end tell -- BBEdit
end open
-- ----------
on cleanUpPageMill(docList)
-- here's the meat of the script!
-- to omit any individual pattern, comment out its subroutine here
eraseWhiteSpace() of me
HRline() of me
elimDoubleSpace() of me -- comment out if using tags
spaceBR() of me
endHeader() of me
orphaned_P() of me
addressBR() of me
double_BR() of me
eraseNATSIZEFLAG() of me
isolateHeadBody() of me
tooManyReturns() of me
end cleanUpPageMill
-- ----------
on checkHTML()
tell application "BBEdit 3.5"
go to line 0
find ""
set HTML_File to the result
find "" without grep
set HTML_File_End to the result
go to line 0
end tell --BBEdit
end checkHTML
-- ----------
on eraseWhiteSpace()
tell application "BBEdit 3.5"
-- Pattern 1
-- this grep pattern erases spaces at the end of a line
-- which PageMill seems to like to include
replace Every Occurrence searching for " *$" using "" with grep
end tell --BBEdit
end eraseWhiteSpace
-- ----------
on HRline()
tell application "BBEdit 3.5"
-- Pattern 2
-- this Find/Replace (with the one following it) places
-- the
tag on its own line
replace Every Occurrence searching for ">
<" using ">
<"
-- ----------
replace Every Occurrence searching for "
" using "
"
end tell --BBEdit
end HRline
-- ----------
on elimDoubleSpace()
tell application "BBEdit 3.5"
-- Pattern 3
-- IMPORTANT!
-- COMMENT THIS PATTERN OUT if you are using the
-- tags, or if for any other reason you want to keep multiple spaces
-- this set of Find/Replaces eliminates two or more spaces
-- in a row; note the grep pattern in the middle "space-space-space-*"
-- which will find two or more spaces in a row.
replace Every Occurrence searching for " " using "%%%%%<TITLE>"
replace Every Occurrence searching for " *" using " " with grep
replace Every Occurrence searching for "%%%%%<TITLE>" using " <TITLE>"
end tell --BBEdit
end elimDoubleSpace
-- ----------
on spaceBR()
tell application "BBEdit 3.5"
-- Pattern 4
-- cleans up unnecessary spaces after periods
replace Every Occurrence searching for " *<BR>
" using "<BR>
" with grep
-- ----------
replace Every Occurrence searching for " *<P>
" using "<P>
" with grep
end tell --BBEdit
end spaceBR
-- ----------
on endHeader()
tell application "BBEdit 3.5"
-- Pattern 5
-- PageMill likes to move the ending header tag to a new line
-- by itself; this grep pattern removes the return character
-- preceeding any ending tag
replace Every Occurrence searching for ">
</H([1-6])>" using "></H\\1>" with grep
end tell --BBEdit
end endHeader
-- ----------
on orphaned_P()
tell application "BBEdit 3.5"
-- Pattern 6
-- adds a blank line before orphaned <P> tags
replace Every Occurrence searching for ">
<P>" using ">
<P>"
end tell --BBEdit
end orphaned_P
-- ----------
on addressBR()
tell application "BBEdit 3.5"
-- Pattern 7
-- I couldn't make PageMaker place more than one <BR> tag before
-- an <ADDRESS> tag, so I replace it with a <P>
replace Every Occurrence searching for "<BR>
<ADDRESS>" using "<P>
<ADDRESS>"
end tell --BBEdit
end addressBR
-- ----------
on double_BR()
tell application "BBEdit 3.5"
-- Pattern 8
-- this Find/Replace eliminates the dreaded "double-<BR>" tags
-- which PageMill always inserts instead of the more correct <P>
replace Every Occurrence searching for "<BR>
<BR>" using "<P>
"
end tell --BBEdit
end double_BR
-- ----------
on eraseNATSIZEFLAG()
tell application "BBEdit 3.5"
-- Pattern 9
-- eliminates the NATURALSIZEFLAG attribute from <IMG> tags
replace Every Occurrence searching for " NATURALSIZEFLAG=
\"[0-3]\"" using "" with grep
replace Every Occurrence searching for "
NATURALSIZEFLAG=\"[0-3]\"" using "" with grep
replace Every Occurrence searching for " NATURALSIZEFLAG
=\"[0-3]\"" using "" with grep
replace Every Occurrence searching for " NATURALSIZEFLAG=\"[0-3]\"
" using "" with grep
replace Every Occurrence searching for " NATURALSIZEFLAG=\"[0-3]\"" using "" with grep
end tell --BBEdit
end eraseNATSIZEFLAG
-- ----------
on isolateHeadBody()
tell application "BBEdit 3.5"
-- Pattern 10
-- adds a line between the Header and Body sections
-- adds a line between the last line of the text and the end
-- of the Body
replace Every Occurrence searching for "</HEAD>
<BODY(.*)>" using "</HEAD>
<BODY\\1>
" with grep
-- ----------
replace Every Occurrence searching for "</BODY>
</HTML>" using "
</BODY>
</HTML>"
end tell --BBEdit
end isolateHeadBody
-- ----------
on tooManyReturns()
tell application "BBEdit 3.5"
-- Pattern 11
-- eliminates extra returns that might have crept in
replace Every Occurrence searching for "
*" using "
" with grep
end tell -- BBEdit
end tooManyReturns