diff options
1 files changed, 27 insertions, 7 deletions
| -rw-r--r-- | generate-hooks/create-description.awk | 34 |
diff --git a/generate-hooks/create-description.awk b/generate-hooks/create-description.awk index b4fac0e..ec0bace 100644 --- a/generate-hooks/create-description.awk +++ b/generate-hooks/create-description.awk @@ -1,18 +1,38 @@ -/p>.*/ { +# For every page, the description meta tag is "[DESCRIPTION]", with this we +# generate it. At most, the first 200 characters from all paragraphs are taken. + +function appToDesc(l) { # Removes tags and bad characters - gsub(/<[^>]*>/, "") ; gsub(/[^a-zA-Z0-9 ]/, "") + gsub(/<[^>]*>/, "", l) ; gsub("’", "'", l) ; gsub(/[^a-zA-Z0-9 '.]/, "", l) # Trims leading and trailing spaces and tabs (if any) - gsub(/^[ \t]*/, "") ; gsub(/[ \t]*$/, "") + gsub(/^[ \t]*/, "", l) ; gsub(/[ \t]*$/, "", l) # Concatenates the current record to variable and shortens it, # when it gets too long - desc = desc $0 + desc = desc l if (length(desc) >= 200) { desc = substr(desc, 1, 197) "..." - exit + return 1 } + return 0 +} + +BEGIN { + stop = 0 + + # Loop over the whole file, extract all text in first couple of paragraphs, + # store it in desc (but no more than 200 characters will be saved) + while(stop == 0 && getline < ARGV[1]) { + if (index($0, "<p") > 0) + stop = appToDesc($0) + } + close(ARGV[1]) # FILENAME isn't set during BEGIN +} + +"\\[DESCRIPTION\\]" { + gsub("\\[DESCRIPTION\\]", desc) } -END { - print desc +{ + print } |