View documentation | View script | License |
Download script | History | Other scripts by: carl |
7-Sep 23:23 UTC
[0.04] 27.947k
[0.04] 27.947k
Archive version of: blog.r ... version: 5 ... carl 7-Jan-2005REBOL [ Title: "REBOL Blogger" Author: "Carl Sassenrath" Version: 1.2.0 File: %blog.r Date: 6-Jan-2005 Purpose: { The blog system written and used by Carl Sassenrath, REBOL's creator. This script will let you post and update blogs on your website using just a web browser. Generates summary and index pages, blog searches, etc. Extensible with Makedoc2 for more formatting options. } Library: [ level: 'intermediate platform: 'all type: [tool] domain: [cgi html http] tested-under: [core 2.5.6 on "Linux x86"] support: none license: 'BSD see-also: none ] ] ; Licensed under new BSD License. To summarize it: ; NO WARRANTY OF ANY KIND. USE ENTIRELY AT YOUR OWN RISK. ; Keep all the above information intact. ; Blog.r Feature Summary: ; 1. Allows remote blogging via html forms. ; 2. Automatically builds main blog page. ; 3. Creates a dated index to all blogs. ; 4. Creates links to and between blogs. ; 5. Use your website's standard page template. ; 6. Supports ranked seaches on blogs. ; 7. Provides a handy blog-content backup method. ; 8. Provides a form for feedback (but feedback system ; is not a standard part of this blog.r script). ; 9. No database engine is needed. ; Installation: ; 1. Upload this script to your CGI script directory. ; 2. Verify that you have a copy of REBOL/Core on your server. ; 3. Add a top line to point to REBOL/Core. Something like: ; #!/usr/bin/rebol -cs ; 4. Change permissions (chmod 755) for blog.r to let it run. ; 5. Test it from a browser. If it has errors, you may need to ; create the blog-dir dir and blog-reads file manually then ; give them proper permissions (chmod 777 or 666). Be careful. ; Read http://www.rebol.com/docs/cgi2.html for details. print "content-type: text/html^/" ; activate web server content output ;-- Configuration ------------------------------------------------------------ ; Location of the blog CGI script file. Note that the script name blog.r ; is found in multiple places in this script, so if you change it here, ; you will also need to search and replace it in the rest of the script. blog-url: http://www.example.com/cgi-bin/blog.r ; Where blogs are stored relative to this script: blog-dir: %blogs/ ; How many blogs are printed on the main page: max-blogs: 5 ; How many blogs are indexed at the bottom of the main page: max-links: 10 ; The blog author/admin IP address. Only this address gets to post. ; If your client uses dynamic IP, then this will not work and you ; will need to use a username/pass with cookies to protect your blog. ; (That might be added to the next version of this blog script.) admin-ip: "100.100.100.100" ; The name of the blogger: author: "Your Name Here" ; IMG source link to author's photo location: author-photo: %/photos/you.jpg ; The name of the organization: organization: "Your Company Here" ; The titles of generated pages: title: context [ main-page: "Your Name's Blog" index-page: "Complete Index of Your Name Blogs:" search-page: "Results of Searching Blog:" feedback-page: "Send Feedback to Your Name" ] ; The meta decscription needed for search engines: meta-description: trim/lines { Comments and ideas provided by Your Name via the REBOL Blogger. } ; Blog page contents can be formatted with makedoc2 allowing nicer ; formatting of examples, etc. If the file below is not found, the ; default formatter will be used (just paragraph separation). makedoc-script: %md2-blog.r ; Automatically detect when running on client in test mode. ; (Causes the tested page to be displayed in the browser.) test-mode: not system/options/cgi/request-method ;-- HTML Templates ----------------------------------------------------------- ; ; If you want to, you can cut and paste these into most HTML page editors to ; help revise them to get what you want. Just be sure to keep the {} braces. ; The name of the HTML template used for the look & feel. Fields within this ; template use the standard $variable format. See show-page function below. template-file: %blog-template.html ; The default HTML template if the above file cannot be found. This also shows ; the variable format needed to create your own blog-template.html file: html-template: { <html> <!--page generated by rebol--> <head><title>$title</title><style> body, p, td {font-family: arial, sans-serif, helvetica; font-size: 10pt;} h1 {font-size: 14pt;} h2 {font-size: 12pt; color: #2030a0; width: 100%; border-bottom: 1px solid #c09060;} h3 {font-size: 10pt; color: #2030a0;} pre, tt {font-family: "courier new", console, monospace; font-size: 10pt;} .title {font-size: 16 pt; font-weight: bold;} .label {margin-top: 3px; margin-bottom: 10px; width: 65px; text-align: right; font-weight: bold; vertical-align: top} </style></head> <body bgcolor="white"> <center> <table width="660" cellpadding="4" cellspacing="0" border="0"> <tr><td> <hr> <h1>$title</h1> <p>$content</p> <hr> </td></tr> <tr><td align=center>$date - $edit - <a href=http://www.rebol.com>Blogger by REBOL</a></td></tr> </table></center></body></html> } ; The HTML boilerplate used at the top of the first page: html-main-boiler: [{ <form action="blog.r" method="get"> <table border=0 cellspacing=1 cellpadding=3> <tr><td width=32></td> <td width=119><img src="} author-photo {"></td><td> <b>} author { <br>} organization { <br>} date/date " " to-GMT date {</b> <br>} blog-count { readers since 4-Jan-2005</b> <br>[<a href="blog.r?index=0">Index of Prior Blogs</a>] <p>Search blogs: <input type=text size=25 name=find> </td></tr></table></form><p> }] ; The HTML boilerplate used at the top of each separate blog page: html-blog-boiler: [{ <form action="blog.r" method="get"> <table border=0 cellspacing=1 cellpadding=3> <tr><td width=32></td><td> <b>} author { <br>} organization { <br>} date/date " " to-GMT date {</b> <br>Blog #} file {</b> <br><span style="color: #808080"> <a href="blog.r">Main page</a> || <a href="blog.r?index=0">Blog index</a>} make-blog-links file { || <a href="blog.r?reply=} file {">Send feedback</a> </span> </td> </tr></table></form><p> }] ; The HTML form used for inputing new blogs: html-edit-form: [{ <form action="blog.r" method="post"> <span class=label>Date:</span> <input type=text size=40 name=date value="} date {"> <br><span class=label>Title:</span> <input type=text size=72 name=title value="} title {"> <br><span class=label>Text:</span> <textarea name=text rows=25 cols=72>} text {</textarea> <br><input type=hidden name=save value=} file {> <br><span class=label> </span> <input type=submit name=submit value=submit> </form> }] ; The HTML form used for feedback replies: html-reply-form: [{ <form action="http://www.example.com/cgi-bin/feedback/post.r" method="post"> <input type="hidden" name="type" size="-1" value="Blog"> <br><span class=label>Subject:</span> <input type=text size=50 name=summary value="} subject {"> <br><span class=label>Message:</span> <textarea name=textarea rows=10 cols=50></textarea> <br><span class=label>Email:</span> <input type=text size=40 name=email value=""> (Optional & Private) <br><span class=label> </span> <input type=submit name=submit value=submit> </form> }] ;-- Utility Functions -------------------------------------------------------- html: "" emit: func [data] [repend html data] safe-html: func [text] [ ; Generate a "safe" html page: foreach [str code] [ "&" "&" "<" "<" ">" ">" ][replace/all text str code] text ] to-GMT: func [date] [ ; Extract a GMT time from DATE either date/time [ date: fourth date - date/zone date/3: 0 ; remove seconds reform [date "GMT"] ][""] ] file-num: func [n] [ ; Output 000n file name format: n: form n insert/dup n "0" 4 - length? n to-file n ] validate-user: does [ ; Validate user has permission to post: if test-mode [return true] if system/options/cgi/remote-addr <> admin-ip [ show-error "Permission denied." ] ] ;-- HTML Page Output --------------------------------------------------------- show-page: func [title /edit link /main /local template tag] [ ; Show page thru CGI or locally in browser (for testing) template: attempt [read template-file] if not template [template: trim/auto html-template] ; use default if main [ tag: build-tag [meta name "description" content (meta-description)] attempt [insert find/tail template <head> tag] ] title: to-string reduce title replace/all template "$title" title if not link [link: 'new] link: rejoin [{<a href="blog.r?edit=} link {">Edit</a>}] replace template "$edit" link replace template "$date" now/date replace template "$content" html either not test-mode [ print template ][ write %temp-page.html template browse %temp-page.html ] quit ] show-error: func [msg] [ ; Output an error page: emit <b> emit msg ; (do not combine) emit </b> show-page "Blogger Error:" ] ;-- Blog File Handling ------------------------------------------------------- blog-obj: context [file: date: title: text: none] load-blog: func [file /local blog] [ attempt [ blog: load/all join blog-dir file blog: construct/with blog blog-obj blog/file: file blog ] ] save-blog: func [file date title text /local blog files] [ validate-user date: to-date date blog: load-blog file if not blog [ files: sort/reverse load blog-dir file: files/1 file: either file [1 + to-integer file] [1] file: file-num file ] save join blog-dir file compose [ date: (date) title: (title) text: (text) ] file ] find-blog: func [blog strs /local rank text][ ; Ranks blog contents during a search: rank: 0 foreach str strs [ if find/any blog/title str [rank: rank + 4] text: blog/text while [text: find/any/tail text str] [rank: rank + 1] ] rank ] backup-blogs: func [start /local files] [ ; This backs up the blog on a remote site so you don't lose them. ; The remote site should CGI with blog.r?back=n where n the start. ; This can easily be done with a few lines of REBOL for backup. validate-user system/options/binary-base: 64 start: to-integer start out: copy [] foreach file load blog-dir [ if (to-integer file) >= start [ repend out [mold file compress read join blog-dir file] ] ] print out quit ] ;-- Blog Page Formatters ----------------------------------------------------- title-blog: func [blog] [ ; Generates a blog title line: reform [blog/date/date "-" blog/title] ] link-blog: func ['cmd value text] [ ; Generates an HTML hyper link: text: to-string reduce text rejoin [{<a href="} blog-url #"?" cmd #"=" value {">} text </a>] ] blog-line: func [blog] [ ; Generates a full blog summary line: rejoin [ "" <span style="text-align: right; width: 1in"> blog/date/date </span> " - " <b> link-blog view blog/file blog/title </b> " [" blog/file "]" newline ] ] format-text: func [text] [ ; Formats text for output: either exists? makedoc-script [ ; Use the Makedoc2 text-to-html formatter: do makedoc-script if not find text "^/^/###" [append text "^/^/###"] second gen-html scan-doc text ][ replace/all text "^/^/" "^/<p>" ] ] show-blog: func [file /local blog date] [ ; Show a single blog page: blog: load-blog file if not blog [show-error ["Blog " file " was not found"]] date: blog/date emit bind html-blog-boiler 'blog emit format-text blog/text show-page/edit blog/title file ] show-main: has [n blog date] [ ; Show the main blog page: date: now emit bind html-main-boiler 'date n: 1 foreach file sort/reverse load blog-dir [ if blog: load-blog file [ either n > max-blogs [ emit [blog-line blog <br>] ][ emit [<h2> title-blog blog " [" link-blog view file file "]" </h2>] emit format-text blog/text ] if n = max-blogs [emit [<h2> "Other Recent Blogs:" </h2><blockquote>]] n: n + 1 if (n - max-blogs) > max-links [break] ] ] emit </blockquote> emit {<b><a href="blog.r?index=0">Click here for list of all prior blogs.</a></b><p>} show-page/main title/main-page ] show-index: has [blog] [ ; Show the blog index page: emit <blockquote> foreach file sort/reverse load blog-dir [ if blog: load-blog file [emit [blog-line blog <br>]] ] emit </blockquote> show-page title/index-page ] show-search: func [text /local rank list blog] [ ; Show the results of a blog search, listed by search-hit rank: text: parse text none list: copy [] foreach file load blog-dir [ if blog: load-blog file [ rank: find-blog blog text if rank > 0 [repend list [rank blog]] ] ] sort/reverse/skip list 2 emit ["The search found " (length? list) / 2 " blogs:<p><blockquote>"] foreach [rank blog] list [emit [blog-line blog <br>]] emit </blockquote> show-page title/search-page ] show-edit: func [file /blog] [ ; Show the blog edit form: validate-user either blog: load-blog file [ emit-edit file blog/date blog/title safe-html blog/text show-page ["Edit blog " file ":"] ][ emit-edit 0 now "" "" show-page "Submit a new blog:" ] ] show-reply: func [blog] [ ; Show the reply form. Note that the reply processing ; script is not part of this blog system. emit-reply reform ["Reply to Blog" blog] show-page title/feedback-page ] make-blog-links: func [file /locals spot out] [ ; Link to prior and next blogs in sequence: out: make string! 30 spot: find load blog-dir to-file file if not spot [return ""] if not head? spot [ file: spot/-1 repend out [" || " link-blog view file ["Prior Blog [" file "]"]] ] if file: spot/2 [ repend out [" || " link-blog view file ["Next Blog [" file "]"]] ] out ] emit-edit: func [file date title text] [ emit bind html-edit-form 'file ] emit-reply: func [subject] [ emit bind html-reply-form 'subject ] ;-- CGI Command Handler ------------------------------------------------------ read-cgi: func [ "Read CGI data. Return data as string or NONE." /limit size "Limit to this number of bytes" /local data buffer ][ if none? limit [size: 300000] switch system/options/cgi/request-method [ "POST" [ data: make string! 1020 buffer: make string! 16380 while [positive? read-io system/ports/input buffer 16380][ append data buffer clear buffer if (length? data) > size [ print ["aborted - posting is too long:" length? data "limit:" size] quit ] ] ] "GET" [data: system/options/cgi/query-string] ] any [data ""] ] ; Possible fields returned by CGI: cgi-obj: context [ view: find: edit: save: back: logon: date: title: text: reply: index: none ] blog-count: 0 ; general blog page hit counter do-cgi: has [cgi] [ blog-count: 1 + any [attempt [load %blog-reads] 0] save %blog-reads blog-count cgi: construct/with decode-cgi read-cgi cgi-obj if cgi/view [show-blog cgi/view] if cgi/index [show-index] if cgi/reply [show-reply cgi/reply] if cgi/edit [show-edit cgi/edit] if cgi/find [show-search cgi/find] if cgi/save [ show-blog save-blog cgi/save cgi/date cgi/title cgi/text ] if cgi/back [backup-blogs cgi/back] show-main ] if not exists? blog-dir [make-dir blog-dir] ;-- Tests ; Uncomment any of the lines below. But, dont' forget to comment it ; back when you upload it back to your server. ;save-blog 0 now/date "Test" "This is a test" ;show-index ;show-main ;backup-blogs 0 ;show-blog %0001 ;show-edit %0001 ;show-search "draw" do-cgi ; start it (if no tests are uncommented) Notes
|