February 2007 - cl-markdown-devel@common-lisp.net

[cl-markdown-devel] div tags
by Hans Halvorson 28 Feb '07

28 Feb '07

cl-markdown does not yet treat <div> tags in the special way that perl-markdown treats them; instead cl-markdown wraps <div> tags in <p> tags. Here is an example input string: input string (9 lines total): ============================= <div class="header"> Header text </div> # Heading Some text Here are the respective output strings: cl-markdown output: =================== <p><div class="header"> </p><p>Header text </p><p></div> </p><h1>Heading </h1> <p>Some text </p> perl-markdown output: =====================x <div class="header"> Header text </div> <h1>Heading</h1> <p>Some text</p> See the description of how perl-markdown treats div tags: http://daringfireball.net/projects/markdown/syntax#html Thanks, Hans

3 4

[cl-markdown-devel] manual line breaking
by Hans Halvorson 27 Feb '07

27 Feb '07

In perl-markdown, a couple of whitespaces at the end of a line are rendered as a <br /> tag. It does not appear that cl-markdown has implemented this feature yet. For example, consider the following input text (note that there are two whitespaces at the end of "First line"): First line Second line Output from perl-markdown: ========================== <p>First line<br /> Second line</p> Output from cl-markdown: ======================== <p>First line Second line</p> Best wishes, Hans

1 0

[cl-markdown-devel] Re: feature request (user guide)
by Gary King 27 Feb '07

27 Feb '07

(cc'd to list -- hope you don't mind) Hi Hans, I'm glad you got them to work and I apologize that they are so _crufty_. I think what I'd like to do is make a macro that does some of the grudge work for you. I have vague hope that I'll do a first step along these lines this week). On Feb 26, 2007, at 4:22 AM, Hans Halvorson wrote: > Dear Gary, > > Thanks for writing the cl-markdown user guide --- it has been very > helpful for me. > > I have one request/suggestion for the user guide: > > When writing my own extensions, I was stymied for a while because I > was getting errors telling me that my function was being called with > the wrong number of arguments. For example, I defined the following > function to give the version number of a system: > > (defun version-number (systemname) > (asdf::component-version (asdf:find-system systemname))) > > If I put "{version-number cl-markdown}" in my document, I got a couple > of errors --- the first error told me that version-number was being > passed the wrong number of arguments. After looking at your examples, > I figured that I should define: > > (defun version-number-2 (phase systemname result) > (declare (ignore phase result)) > (asdf::component-version (asdf:find-system systemname)))) > > I then received an error that "(cl-markdown) isn't a valid system > name." So, I defined: > > (defun version-number-3 (phase systemname result) > (declare (ignore phase result)) > (version-number (car systemname))) > > and it worked. So, could you write something in the user guide > explaining the proper format for extension functions? > > Thanks, > Hans -- Gary Warren King, metabang.com Cell: (413) 885 9127 Fax: (206) 338-4052 gwkkwg on Skype * garethsan on AIM

1 0

[cl-markdown-devel] Re: cl-markdown (hangs head in shame)
by Gary King 21 Feb '07

21 Feb '07

(added the mailing list) Hi Luke, No apologies necessary; I'm sorry that this is taking so much effort on your part and appreciate that you are sticking with it. I'll look into the problem you are seeing with defsystem- compatibility. FWIW, I just successfully loaded the bundled version (currently at http://www.metabang.com/resources/cl-markdown- bundle.tar.gz) in the LispWorks personal edition. So if you haven't tried that yet, it might be worth doing so. thanks again, On Feb 20, 2007, at 11:34 PM, Luke J Crook wrote: > Hi Gary, > I must appologize for not trying cl-markdown after you made your > fixes last week. I know it sucks when I make changes in lispbuilder > per user feedback and then don't receive further feedback. > Anyway, I tried cl-markdown today with the latest versions of > everything using darcs pull. > The bad news is that cl-markdown still fails. The error returned in > the SLIME buffer not make much sense, however the error in the > Lipsworks buffer seems to shed more light. I have included the > Lispworks buffer first and then the SLIME buffer after. > The versions of the packages that I am using are as follows: > - asdf-system-connections 0.8.3 > - cl-containers 0.8.3 > - cl-fad 0.5.1 > - cl-markdown 0.8.6 > - cl-ppcre 1.2.19 > - cl-who 0.6.3 > - closer-mop (Feb 06 version) > - defsystem-compatibility (Fri Jun 23 version) > - html-encode 1.2 > - lw-compat 0.22 > - lw-doc 0.1.5 > - metabang-bind (Feb 5 version) > - metatiltiies 0.6.3 > - moptilities (0.3.4) > > - Luke >>>>> Start Lispworks buffer error > > ;;; Compiling file c:\home\bin\clisp\asdf\cl-markdown\dev > \package.lisp ... > ;;; Safety = 3, Speed = 1, Space = 1, Float = 1, Interruptible = 0 > ;;; Compilation speed = 1, Debug = 2, Fixnum safety = 3 > ;;; Source level debugging is on > ;;; Source file recording is on > ;;; Cross referencing is on > ; (TOP-LEVEL-FORM 1) > ; (TOP-LEVEL-FORM 2) > > **++++ Error in (PACKAGE "CL-MARKDOWN"): > No symbol named "SYSTEM-RELATIVE-PATHNAME" is accessible in > #:DEFSYSTEM-COMPATIBILITY. > > **++++ Error in (PACKAGE "CL-MARKDOWN-USER"): > Package CL-MARKDOWN not found. > ; (TOP-LEVEL-FORM 3) > ; *** 2 errors detected, no fasl file produced. > <<<<< End Lispworks buffer error > >>>>>> Start SLIME buffer error > > erred while invoking #<ASDF:COMPILE-OP NIL 21EB9EE4> on > #<ASDF:CL-SOURCE-FILE "package" 21EBFC7C> > [Condition of type ASDF:COMPILE-ERROR] > Restarts: > 0: [RETRY] Retry performing #<ASDF:COMPILE-OP NIL 21EB9EE4> on > #<ASDF:CL-SOURCE-FILE "package" 21EBFC7C>. > 1: [ACCEPT] Continue, treating #<ASDF:COMPILE-OP NIL 21EB9EE4> on > #<ASDF:CL-SOURCE-FILE "package" 21EBFC7C> as having been successful. > 2: [ABORT-REQUEST] Abort handling SLIME request. > 3: [ABORT] Quit process. > Backtrace: > 0: CONDITIONS::CONDITIONS-ERROR (:INVISIBLEP T ASDF:COMPILE-ERROR > (:COMPONENT #<ASDF:CL-SOURCE-FILE "package" 21EBFC7C> :OPERATION > #<ASDF:COMPILE-OP NIL 21EB9EE4>)) > 1: ("Anonymous compiled function ~s" #<function 203E5E4A>) NIL > 2: (SUBFUNCTION 1 ASDF:OPERATE) NIL > 3: COMPILER::DO-COMPILATION-UNIT (#<closure (SUBFUNCTION 1 > ASDF:OPERATE) 21EC17A2> T NIL) > 4: COMPILER::WITH-COMPILATION-UNIT-INTERNAL (#<closure > (SUBFUNCTION 1 ASDF:OPERATE) 21EC17A2> &KEY NIL NIL NIL) > 5: ASDF:OPERATE (ASDF:LOAD-OP :CL-MARKDOWN &REST NIL &KEY T NIL > "<&ALLOW-OTHER-KEYS>") > 6: SPECIAL::%EVAL-NOHOOK ((ASDF:OPERATE 'ASDF:LOAD-OP :CL-MARKDOWN)) > 7: IV:PROCESS-TOP-LEVEL (DSPEC:*CURRENT-FORM* #S > (LEXICAL::ENVIRONMENT :VARIABLES NIL :FUNCTIONS NIL :REMOTE- > ENVIRONMENT NIL) #<function LEXICAL:EVAL-WITH-ENV 20140F9A> > &OPTIONAL :EXECUTE) > 8: SWANK::EVAL-REGION ("(asdf:operate 'asdf:load-op :cl-markdown) > " > &OPTIONAL > T) > 9: (SUBFUNCTION 1 SWANK:LISTENER-EVAL) NIL > 10: SWANK::CALL-WITH-BUFFER-SYNTAX (#<closure (SUBFUNCTION 1 > SWANK:LISTENER-EVAL) 21E7524A>) > 11: SPECIAL::%EVAL-NOHOOK ((SWANK:LISTENER-EVAL "(asdf:operate > 'asdf:load-op :cl-markdown) > ")) > 12: IV:PROCESS-TOP-LEVEL (DSPEC:*CURRENT-FORM* #S > (LEXICAL::ENVIRONMENT :VARIABLES NIL :FUNCTIONS NIL :REMOTE- > ENVIRONMENT NIL) #<function LEXICAL:EVAL-WITH-ENV 20140F9A> > &OPTIONAL :EXECUTE) > 13: (SUBFUNCTION 1 SWANK::EVAL-FOR-EMACS) NIL > 14: (SUBFUNCTION 1 (TOP-LEVEL-FORM 52)) (#<function SWANK:SWANK- > DEBUGGER-HOOK 21FFD6E2> #<closure (SUBFUNCTION 1 SWANK::EVAL-FOR- > EMACS) 21E75312>) > 15: FUNCALL (SWANK::EVAL-FOR-EMACS &REST > ((SWANK:LISTENER-EVAL "(asdf:operate > 'asdf:load-op :cl-markdown) > ") > "COMMON-LISP-USER" > 7)) > <<<<< End SLIME buffer error -- Gary Warren King, metabang.com Cell: (413) 885 9127 Fax: (206) 338-4052 gwkkwg on Skype * garethsan on AIM

1 0

[cl-markdown-devel] announce: CL-Markdown 0.8.5
by Gary King 06 Feb '07

06 Feb '07

Today's update to [CL-Markdown][] marks brings more fixes to paragraphs and entity encoding. It also finds CL-Markdown saying goodbye to [HTML-Encode][] (because I decided that because I only used a tiny bit of it and because "less dependencies" is a good thing, it was better just to pull the bit out and make it my own...). Oh, there's one more thing: if you're generating HTML (and, frankly, who isn't nowadays), then CL-Markdown looks for the following properties: * author * description * copyright * keywords * date * refresh * expires If it finds one, it adds the appropriate META tag to the HTML header (i.e., either `name` or `http-equiv`). There is no sanity checking but it should handle the basics. If it doesn't, let me know. [HTML-Encode]: http://www.cliki.net/html-encode [CL-Markdown]: http://common-lisp.net/project/cl-markdown/ [Trivial-shell]: http://common-lisp.net/project/trivial-shell/ [comparison]: http://common-lisp.net/project/cl-markdown/comparison- tests/ -- Gary Warren King, metabang.com Cell: (413) 885 9127 Fax: (206) 338-4052 gwkkwg on Skype * garethsan on AIM

1 0

[cl-markdown-devel] author property broken?
by Hans Halvorson 05 Feb '07

05 Feb '07

I am running tests with 0.8.4, and the {set-property Author ...} seems to be malfunctioning. For example: (let ((*parse-active-functions* '(table-of-contents set-property)) (*render-active-functions* '(table-of-contents property))) (markdown "{set-property html t} {set-property title My best summer vacation ever} {set-property style-sheet my-style.css} {set-property Author Gary W. King} Some text." :stream nil :format :html)) => #<DOCUMENT {A7D72B9}> "<!DOCTYPE html PUBLIC \"-//W3C//DTD XHTML 1.0 Strict//EN\" \"http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd\"> <html> <head> <title>My best summer vacation ever</title> <link type='text/css' href='my-style.css' rel='stylesheet' /> </head> <body><p> Some text. </p> </body> </html>" Cheers, Hans

1 1

[cl-markdown-devel] announcing cl-markdown 0.8.4
by Gary King 05 Feb '07

05 Feb '07

Thanks to some more bug finding by Hans Halvorson, [CL-Markdown][] now correctly lower-cases paragraph markers and handles hard returns more correctly. I've also extended the test-suite (using the new features of [Trivial-shell][]) so that it's easier to test compatibility between markdown and CL-Markdown. To the end, I've also re-generated the [comparison][] page; There are a few new problems (cf. auto links and the syntax-compare!) so I'll hope to get to 0.8.5 very soon. [CL-Markdown]: http://common-lisp.net/project/cl-markdown/ [Trivial-shell]: http://common-lisp.net/project/trivial-shell/ [comparison]: http://common-lisp.net/project/cl-markdown/comparison- tests/ -- Gary Warren King, metabang.com Cell: (413) 885 9127 Fax: (206) 338-4052 gwkkwg on Skype * garethsan on AIM

1 0

[cl-markdown-devel] list items with newlines
by Hans Halvorson 03 Feb '07

03 Feb '07

(Perl)-markdown and cl-markdown give different outputs for the following input: * A first list item with a hard return * A second list item The perl-markdown output: <ul> <li>A first list item with a hard return</li> <li>A second list item</li> </ul> The cl-markdown output: <ul> <li>A first list item</li> </ul>with a hard return <ul> <li>A second list item</li> </ul> So, cl-markdown reads the newline followed by a line not beginning with "*" as indicating that the list item and list are ending. For further diagnosis, I did: (cl-containers:contents (chunks (chunk-source "* A first list item with a hard return * A second list item"))) => #(#<CHUNK NIL/1 1 lines LINE-STARTS-WITH-BULLET-P LINE-IS-NOT-EMPTY-P> #<CHUNK NIL/0 1 lines LINE-IS-NOT-EMPTY-P LINE-STARTS-WITH-BULLET-P> #<CHUNK NIL/1 1 lines LINE-STARTS-WITH-BULLET-P END-OF-DOCUMENT>) But at this point, I can't tell what is happening: is the second chunk (line) being misread? Or is this OK, and the problem occurs later, e.g. when the first and second chunks should be merged? Thanks, Hans

2 1

[cl-markdown-devel] links, lists, headings
by Hans Halvorson 02 Feb '07

02 Feb '07

I have come across some weird behavior in the interaction between links, lists, and headings (cl-markdown v 0.8.2). The following two forms act as they should: (markdown "## Common Lisp * An item with a link [link](link.html) and some following text." :format :html :stream nil) => "<H2>Common Lisp </H2> <UL><LI>An item with a link <a href=\"link.html\">link</a> and some following text. </LI> </UL>" (markdown "## Common Lisp * An item with a link [link](link.html) and some following text. * Another item" :format :html :stream nil) => "<H2>Common Lisp </H2> <UL><LI>An item with a link <a href=\"link.html\">link</a> and some following text. </LI> <LI>Another item </LI> </UL>" But the following form wraps the first list item in paragraph tags: (markdown "## Common Lisp * An item with a link [link](link.html) and some following text. ## A second level heading * Another item" :format :html :stream nil) => "<H2>Common Lisp </H2> <UL><LI><P>An item with a link <a href=\"link.html\">link</a> and some following text. </P></LI> </UL><H2>A second level heading </H2> <UL><LI>Another item </LI> </UL>" I also compared the output for this last string with the output from Perl-markdown, and it does *not* add the paragraph tags. By the way, shouldn't the xhtml tags be lower case by default? Thanks, Hans

2 3

[cl-markdown-devel] Announce 0.8.3
by Gary King 02 Feb '07

02 Feb '07

Thanks to Hans Halvorson for noticing both that [CL-Markdown][] wasn't handling list item paragraph tags properly *and* that it was generating HTML tag in ALL UPPERCASE. Both of these items are fixed in version 0.8.3 ([changelog][]). [CL-Markdown]: http://common-lisp.net/project/cl-markdown/ [changelog]: http://common-lisp.net/project/cl-markdown/changelog.html -- Gary Warren King, metabang.com Cell: (413) 885 9127 Fax: (206) 338-4052 gwkkwg on Skype * garethsan on AIM

1 0