On Oct 19, 2020, at 1:40 PM, w...@apache.org wrote:
Author: wave
Date: Mon Oct 19 20:40:10 2020
New Revision: 1882669
URL: http://svn.apache.org/viewvc?rev=1882669&view=rev
Log:
Further analysis of current templates. Analysis of possible frameworks.
Added:
openoffice/ooo-site/trunk/templates/FRAMEWORKS.txt
Modified:
openoffice/ooo-site/trunk/templates/README.txt
openoffice/ooo-site/trunk/templates/ssi.json
Added: openoffice/ooo-site/trunk/templates/FRAMEWORKS.txt
URL:
http://svn.apache.org/viewvc/openoffice/ooo-site/trunk/templates/FRAMEWORKS.txt?rev=1882669&view=auto
==============================================================================
--- openoffice/ooo-site/trunk/templates/FRAMEWORKS.txt (added)
+++ openoffice/ooo-site/trunk/templates/FRAMEWORKS.txt Mon Oct 19 20:40:10 2020
@@ -0,0 +1,30 @@
+Observations about the suitability and challenges using various frameworks.
+
+(1) Pelican
+- Suitable for our Markdown, but we will need to rename all files - mdtext ->
md
+- Has it's own methods for HTML which are not really compatible with our
current methods.
+ We would need to write our own plugin and rename all of the htm and html
files to a special prefix like aoo.
+
+Seems like a bad fit. Thumbs down.
+
+(2) JBake
+- Suitable for our Markdown, but will need to rename all files - mdtext -> md
+- Html and Htm will need to have a minimal piece of metadata inserted in order
to call out the particular template.
+ See https://jbake.org/docs/2.6.5/#content_files and
https://jbake.org/docs/2.6.5/#custom_templates
+
+- Assets and Content is separated so that all of our as is files should be in
a sepearate tree.
+- We can obsolete directories without deleting the content via
https://jbake.org/docs/2.6.5/#ignoring_files_and_directories
+
+This could be a good fit. Thumbs up.
+
+(3) Hugo
+- Go based.
+- Looks like a lot of rewrite.
+
+No opinion, but it looks like a steep learning curve
+
+(4) Jekyll
+- Ruby based
+- Suitable for our Markdown, but will need to rename all files - mdtext -> md
+- Unclear how we would handle the html parsing.
+
Modified: openoffice/ooo-site/trunk/templates/README.txt
URL:
http://svn.apache.org/viewvc/openoffice/ooo-site/trunk/templates/README.txt?rev=1882669&r1=1882668&r2=1882669&view=diff
==============================================================================
--- openoffice/ooo-site/trunk/templates/README.txt (original)
+++ openoffice/ooo-site/trunk/templates/README.txt Mon Oct 19 20:40:10 2020
@@ -5,15 +5,24 @@ Some thoughts and documentation.
(2) SSI directory structure could be replaced could be replaced by a json file.
- footer is always the same.
+ Path to the footer file to include. Always the same.
- doctype is always the same.
-- home is always the same. and is not used.
+ Path to the doctype file to use. Ought to eliminate this hack.
+- home is always the same.
+ The name to use for the first breadcrumb which will take you to "/".
- brand for the top of the page.
+ Path to the brand file to include. Varies according the language site.
- topnav for the upper navigation bar.
+ Path to the top nav menu to include. Varies according the language site.
- leftnav is optional and special.
+ Path to a left nav menu to include. Varies according the language site or
"project" site.
- rightnav is optional and only used once for l10n where there is also a
leftnav.
+ Path to a right nav menu to include. Varies according the language site or
"project" site.
(3) ssi.json is a directory of directories.
+Note this actitivty should also build the breadrcumb list. See (5)
+
For each template key look up the subpath and then look for the ssi element
name.
If missing then go up a level on the subpath.
@@ -30,7 +39,8 @@ For api:
"doctype":"/doctype.html",
"footer":"/footer.html",
"brand":"/brand.html",
- "topnav":"/topnav.html"
+ "topnav":"/topnav.html",
+ "home":"home"
}
@@ -46,7 +56,8 @@ For de:
"doctype":"/doctype.html",
"footer":"/footer.html",
"brand":"/brand.html",
- "topnav":"/topnav.html"
+ "topnav":"/topnav.html",
+ "home":"home"
}
For xx:
@@ -64,11 +75,14 @@ For xx:
"doctype":"/doctype.html",
"footer":"/footer.html",
"brand":"/brand.html",
- "topnav":"/topnav.html"
+ "topnav":"/topnav.html",
+ "home":"home"
}
(4) List of templates:
+See lib/path.pm
+
- brand.html
- brand.mdtext
- footer.html
@@ -120,3 +134,78 @@ For xx:
<!--#include virtual="{{ ssi.headers.footer }}" -->
</body>
</html>
+
+(5) Breadcrumbs
+
+sub breadcrumbs {
+ my @path = split m!/!, shift;
+ pop @path;
+ my @rv;
+ my $relpath = "";
+ for (@path) {
+ $relpath .= "$_/";
+ $_ ||= "home";
+ push @rv, qq(<a href="$relpath">$_</a>);
+ }
+ return join " » ", @rv;
+}
+
+Consider converting into an ul / li structure
+
+Combine with ssi processing.
+
+(6) HTML file rewriting.
+
+In lib/view.pm see the html_page function:
+
+The following match is done on the html source:
+
+ if ($args{content} =~
m!<head.*?>(.*?)</head>(?:.*?<body(.*?)>)?(.*?)(?:</body>|\Z)!si) {
+ @args{qw/head bodytag content/} = ($1, $2, $3);
+ }
+
+Head attributes are ignored.
+All head content is copied.
+Any body attributes are preserved.
+All of the body content is copied.
+
+If a python framework is used then suggest that re.find be used.
+
+(7) CSS
+
+/css/ooo.css - Skeleton css
+/css/exceptions.css - Original exceptions - Home page and downloads
+/css/home.css - Home page
+/css/styles.css - original css - Home page and Downloads
+
+Lots of css scattered throughout the site.
+
+Navigator inserted css which seems to be worthless.
+<style type="text/css">
+/* The following code is added by mdx_elementid.py
+ It was originally lifted from http://subversion.apache.org/style/site.css */
+/*
+ * Hide class="elementid-permalink", except when an enclosing heading
+ * has the :hover property.
+ */
+.headerlink, .elementid-permalink {
+ visibility: hidden;
+}
+h2:hover > .headerlink, h3:hover > .headerlink, h1:hover > .headerlink, h6:hover >
.headerlink, h4:hover > .headerlink, h5:hover > .headerlink, dt:hover > .elementid-permalink {
visibility: visible }</style>
+
+(8) Javascript
+
+/scripts/
+api.simile-widgets.org - used for stats time plots.
+entourage.js - ssi included tracking downloads somehow.
+google-analytics.js - ssi included
+jquery-1.11.2.min.js - upgrade?
+ooo.js
+sorttable.js - other copies used
+tableDecorate.js - other copies used
+
+scripts are not optimized and not really normalized.
+ssi including entourage.js on all pages is a little silly. It only makes sense
on download pages.
+
+We need to determine google-analytics vs. ASF foundation policy.
+
Modified: openoffice/ooo-site/trunk/templates/ssi.json
URL:
http://svn.apache.org/viewvc/openoffice/ooo-site/trunk/templates/ssi.json?rev=1882669&r1=1882668&r2=1882669&view=diff
==============================================================================
--- openoffice/ooo-site/trunk/templates/ssi.json (original)
+++ openoffice/ooo-site/trunk/templates/ssi.json Mon Oct 19 20:40:10 2020
@@ -3,7 +3,8 @@
"doctype":"/doctype.html",
"footer":"/footer.html",
"brand":"/brand.html",
- "topnav":"/topnav.html"
+ "topnav":"/topnav.html",
+ "home":"home"
}
"/sl/": {
"brand":"/sl/brand.html",
