markdownr 0.5.6 → 0.5.8

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 6e04e007ceb9d03b7cadc1cfbb9b7d6d4df51f83d82fecba4fc22d52623b50fe
-  data.tar.gz: ed3c37ef77810c9cd2589314d9d3239fb77f48e0ac3e1ede670146db9ff2c1bc
+  metadata.gz: 44c223ecfa0f81c35d2d3d89a5c316eb2adc7f0a343db241185dc828268d3527
+  data.tar.gz: 7b87a1fed84d540c89bc92bea8de754eee3d680ba93f952ad4b23c7150aa06b3
 SHA512:
-  metadata.gz: dc882e485d1eadb2415443b5056dca89534fe1b561118d508060b126b53332a507ee2ef375faa648444cdc0ef394179de172acdee13981d30d8f89737dba6bed
-  data.tar.gz: 0feb715779946a7456559618e3493986520dcbbeeab7714caa7049e6d4d196a90471fbd90102e9b060ddee7c9f2028fbf6ab5215e9de85c0b793ec4d9776e18c
+  metadata.gz: 8eb259d062a13af47d3126ea380777e01aba1f1e6bc9401fb2f05c8277e9a827d6e16f9299e96768236574114875c0f4d2b5ab273094513c71670afa765f3e12
+  data.tar.gz: 56fda7d165fecbec5b46d8a496d257aa38164684779225bfac6e1d9da00df3fb8da3e90fbd4c58c435f2680a21869271f06c4b3791df48b9ac1b6d64e7359277

data/lib/markdown_server/app.rb

@@ -580,6 +580,145 @@ module MarkdownServer
         out.length > 10_000 ? out[0, 10_000] : out
       end
 
+      def blueletterbible_html(html, url)
+        base = "https://www.blueletterbible.org"
+
+        # ── Word ──────────────────────────────────────────────────────────────
+        word = html[/<h6[^>]+class="lexTitle(?:Gk|Hb)"[^>]*>(.*?)<\/h6>/im, 1]
+                 &.gsub(/<[^>]+>/, "")&.strip || ""
+
+        # ── Transliteration ───────────────────────────────────────────────────
+        transliteration = html[/<div[^>]+id="lexTrans".*?<em>(.*?)<\/em>/im, 1]&.strip || ""
+
+        # ── Pronunciation + audio ─────────────────────────────────────────────
+        pronunciation = html[/class="[^"]*lexicon-pronunc[^"]*"[^>]*>\s*([^\n<]{1,50})/i, 1]&.strip || ""
+        data_pronunc  = html[/data-pronunc="([a-fA-F0-9]{20,})"/i, 1] || ""
+        audio_btn = if data_pronunc.length > 10
+          au = "#{base}/lang/lexicon/lexPronouncePlayer.cfm?skin=#{data_pronunc}"
+          %(<button onclick="var a=this._a||(this._a=new Audio('#{h(au)}'));a.currentTime=0;a.play();" ) +
+            %(style="background:none;border:none;cursor:pointer;padding:0 0 0 4px;font-size:1.1em;vertical-align:middle;" title="Play pronunciation">&#128266;</button>)
+        else
+          ""
+        end
+
+        # ── Part of speech ────────────────────────────────────────────────────
+        pos = html[/<div[^>]+id="lexPart".*?small-text-right"[^>]*>(.*?)<\/div>/im, 1]
+               &.gsub(/<[^>]+>/, "")&.strip || ""
+
+        # ── Info table ────────────────────────────────────────────────────────
+        info_rows = [
+          ["Word",            h(word)],
+          ["Transliteration", "<em>#{h(transliteration)}</em>"],
+          ["Pronunciation",   "#{h(pronunciation)}#{audio_btn}"],
+          ["Part of Speech",  h(pos)],
+        ]
+        info_html = %(<table class="blb-table">) +
+          info_rows.map { |label, v|
+            %(<tr><th class="blb-th">#{h(label)}</th><td>#{v}</td></tr>)
+          }.join + "</table>"
+
+        # ── Inflections ───────────────────────────────────────────────────────
+        infl_html = ""
+        if (m = html.match(/<div\s[^>]*id="greek-tr-inflections"[^>]*>/im))
+          after_infl = html[m.end(0)..]
+          stop       = after_infl.index(/<div\s[^>]*id="greek-(?:mgnt|lxx)-inflections"/i) || after_infl.length
+          infl_section = after_infl[0...stop]
+          inflections  = []
+          infl_section.scan(/<div\s[^>]*class="greekInflection"[^>]*>(.*?)<\/div>\s*<\/div>/im) do |mv|
+            chunk = mv[0]
+            href  = chunk[/href="([^"]+)"/i, 1]
+            gk    = chunk[/<span[^>]+class="Gk"[^>]*>(.*?)<\/span>/im, 1]&.gsub(/<[^>]+>/, "")&.strip || ""
+            freq  = chunk[/&#8212;\s*(\d+)x<\/a>/i, 1]&.to_i || 0
+            next if gk.empty? || freq.zero?
+            inflections << { word: gk, freq: freq,
+                             href: href ? base + href.gsub("&amp;", "&") : nil }
+          end
+          inflections.sort_by! { |i| -i[:freq] }
+          if inflections.any?
+            rows = inflections.map { |i|
+              link = i[:href] ? %(<a href="#{h(i[:href])}" target="_blank" rel="noopener">#{h(i[:word])}</a>) : h(i[:word])
+              %(<tr><td>#{link}</td><td class="blb-right">#{i[:freq]}x</td></tr>)
+            }.join
+            infl_html = %(<h4 class="blb-heading">Inflections</h4>) +
+              %(<table class="blb-table"><thead><tr><th class="blb-th">Form</th>) +
+              %(<th class="blb-th blb-right">Count</th></tr></thead><tbody>#{rows}</tbody></table>)
+          end
+        end
+
+        # ── Biblical Usage ────────────────────────────────────────────────────
+        usage_html = ""
+        if (um = html.match(/<div[^>]+id="outlineBiblical"[^>]*>/im))
+          after_usage = html[um.end(0)..]
+          if (inner = after_usage.match(/\A\s*<div>([\s\S]*?)<\/div>\s*<\/div>/im))
+            cleaned = inner[1]
+              .gsub(/<(\/?)(\w+)[^>]*>/) { "<#{$1}#{$2.downcase}>" }
+              .gsub(/[ \t]+/, " ")
+              .strip
+            usage_html = %(<h4 class="blb-heading">Biblical Usage</h4><div class="blb-usage">#{cleaned}</div>)
+          end
+        end
+
+        # ── Concordance ───────────────────────────────────────────────────────
+        conc_html  = ""
+        trans_name = html[/id="bibleTable"[^>]+data-translation="([^"]+)"/i, 1] || ""
+        verses = []
+        html.split(/<div\s[^>]*id="bVerse_\d+"[^>]*>/).drop(1).each do |chunk|
+          cite_href = chunk[/tablet-order-2[^>]*>[\s\S]{0,400}?href="([^"]+)"/im, 1] || ""
+          cite      = chunk[/tablet-order-2[^>]*>[\s\S]{0,400}?<a[^>]*>(.*?)<\/a>/im, 1]
+                        &.gsub(/<[^>]+>/, "")&.strip || ""
+
+          # Process verse HTML: highlight the matched word, strip all Strong's refs
+          raw_html = chunk[/class="EngBibleText[^"]*"[^>]*>([\s\S]*?)<\/div>/im, 1] || ""
+          raw_html.gsub!(/<img[^>]*>/, "")
+          raw_html.gsub!(/<a[^>]*class="hide-for-tablet"[^>]*>[\s\S]*?<\/a>/im, "")
+          raw_html.gsub!(/<span[^>]*class="hide-for-tablet"[^>]*>[\s\S]*?<\/span>/im, "")
+          # Use control-char placeholders so blb-match survives the tag-strip pass
+          verse_html = raw_html.gsub(/<span\s[^>]*class="word-phrase"[^>]*>([\s\S]*?)<\/span>/im) do
+            inner = $1
+            word = inner.sub(/<sup[\s\S]*/im, "").gsub(/<[^>]+>/, "")
+                        .gsub(/&nbsp;/i, " ").strip
+            inner.match?(/<sup[^>]*class="[^"]*strongs criteria[^"]*"/i) ?
+              "\x02#{word}\x03" : word
+          end
+          # Fallback for translations without word-phrase spans (NASB, ESV, etc.)
+          # The criteria word appears directly before its <sup class="strongs criteria"> tag
+          unless verse_html.include?("\x02")
+            verse_html.gsub!(/([\w]+[,;:.!?'"]*)\s*<sup[^>]*class="[^"]*strongs criteria[^"]*"[\s\S]*?<\/sup>/im) do
+              "\x02#{$1}\x03"
+            end
+          end
+          verse_html.gsub!(/<sup[^>]*>[\s\S]*?<\/sup>/im, "")
+          verse_html.gsub!(/<[^>]+>/, "")
+          verse_html.gsub!(/&nbsp;/i, " ")
+          verse_html.gsub!(/&#(\d+);/)         { [$1.to_i].pack("U")      rescue " " }
+          verse_html.gsub!(/&#x([\da-f]+);/i)  { [$1.to_i(16)].pack("U") rescue " " }
+          verse_html.gsub!(/&amp;/, "&")
+          verse_html.gsub!(/&lt;/, "<")
+          verse_html.gsub!(/&gt;/, ">")
+          verse_html.gsub!(/\s+/, " ")
+          verse_html.strip!
+          # Strip the mobile citation prefix ("Mat 5:17 - ") left by hide-for-tablet removal
+          verse_html.sub!(/\A#{Regexp.escape(cite)}\s*-\s*/i, "")
+          # Restore match placeholders as highlighted spans
+          verse_html.gsub!(/\x02([^\x03]*)\x03/) { %(<span class="blb-match">#{h($1.strip)}</span>) }
+
+          next if cite.empty? || verse_html.empty?
+          full_href = cite_href.empty? ? nil : (cite_href.start_with?("http") ? cite_href : base + cite_href)
+          verses << { cite: cite, verse_html: verse_html, href: full_href }
+        end
+        if verses.any?
+          heading = trans_name.empty? ? "Concordance" : "Concordance (#{h(trans_name)})"
+          rows = verses.map { |v|
+            link = v[:href] ? %(<a href="#{h(v[:href])}" target="_blank" rel="noopener">#{h(v[:cite])}</a>) : h(v[:cite])
+            %(<tr><td class="blb-nowrap">#{link}</td><td>#{v[:verse_html]}</td></tr>)
+          }.join
+          conc_html = %(<h4 class="blb-heading">#{heading}</h4>) +
+            %(<table class="blb-table"><tbody>#{rows}</tbody></table>)
+        end
+
+        info_html + infl_html + usage_html + conc_html
+      end
+
       def compile_regexes(query)
         words = query.split(/\s+/).reject(&:empty?)
         return nil if words.empty?
@@ -622,6 +761,98 @@ module MarkdownServer
       end
     end
 
+    get "/debug/raw-fetch" do
+      url = params[:url].to_s.strip
+      halt 400, "missing ?url=" if url.empty?
+      html = fetch_external_page(url)
+      halt 502, "fetch failed" unless html
+      content_type :text
+      # Show processing steps for first verse
+      chunk = html.split(/<div\s[^>]*id="bVerse_\d+"[^>]*>/).drop(1).first
+      return "no bVerse chunks found" unless chunk
+
+      cite = chunk[/tablet-order-2[^>]*>[\s\S]{0,400}?<a[^>]*>(.*?)<\/a>/im, 1]
+               &.gsub(/<[^>]+>/, "")&.strip || "?"
+      raw_html = chunk[/class="EngBibleText[^"]*"[^>]*>([\s\S]*?)<\/div>/im, 1] || "(no EngBibleText found)"
+
+      lines = ["=== cite: #{cite} ===",
+               "=== EngBibleText raw (#{raw_html.length} chars) ===",
+               raw_html, ""]
+
+      # Simulate the processing steps
+      rh = raw_html.dup
+      rh.gsub!(/<img[^>]*>/, "")
+      rh.gsub!(/<a[^>]*class="hide-for-tablet"[^>]*>[\s\S]*?<\/a>/im, "")
+      rh.gsub!(/<span[^>]*class="hide-for-tablet"[^>]*>[\s\S]*?<\/span>/im, "")
+
+      wp_matches = rh.scan(/<span\s[^>]*class="word-phrase"[^>]*>([\s\S]*?)<\/span>/im)
+      lines << "=== word-phrase matches (#{wp_matches.length}) ==="
+      wp_matches.each_with_index do |(inner), i|
+        is_criteria = inner.match?(/<sup[^>]*class="[^"]*strongs criteria[^"]*"/i)
+        word = inner.sub(/<sup[\s\S]*/im, "").gsub(/<[^>]+>/, "").gsub(/&nbsp;/i, " ").strip
+        lines << "  [#{i}] criteria=#{is_criteria} word=#{word.inspect}"
+      end
+
+      # Now simulate the full processing pipeline
+      verse_html = rh.gsub(/<span\s[^>]*class="word-phrase"[^>]*>([\s\S]*?)<\/span>/im) do
+        inner = $1
+        word = inner.sub(/<sup[\s\S]*/im, "").gsub(/<[^>]+>/, "").gsub(/&nbsp;/i, " ").strip
+        inner.match?(/<sup[^>]*class="[^"]*strongs criteria[^"]*"/i) ? "\x02#{word}\x03" : word
+      end
+      lines << "\n=== after word-phrase gsub (placeholder check) ==="
+      lines << "  contains \\x02: #{verse_html.include?("\x02")}"
+      lines << "  contains \\x03: #{verse_html.include?("\x03")}"
+      ph = verse_html[/\x02[^\x03]*\x03/]
+      lines << "  placeholder found: #{ph.inspect}"
+
+      verse_html.gsub!(/<sup[^>]*>[\s\S]*?<\/sup>/im, "")
+      verse_html.gsub!(/<[^>]+>/, "")
+      verse_html.gsub!(/&nbsp;/i, " ")
+      verse_html.gsub!(/&#(\d+);/)         { [$1.to_i].pack("U") rescue " " }
+      verse_html.gsub!(/&#x([\da-f]+);/i)  { [$1.to_i(16)].pack("U") rescue " " }
+      verse_html.gsub!(/&amp;/, "&")
+      verse_html.gsub!(/&lt;/, "<")
+      verse_html.gsub!(/&gt;/, ">")
+      verse_html.gsub!(/\s+/, " ")
+      verse_html.strip!
+
+      lines << "=== after tag-strip (placeholder check) ==="
+      lines << "  contains \\x02: #{verse_html.include?("\x02")}"
+      ph2 = verse_html[/\x02[^\x03]*\x03/]
+      lines << "  placeholder found: #{ph2.inspect}"
+      lines << "  verse_html snippet: #{verse_html[0, 200].inspect}"
+
+      # Apply the final restore
+      restored = verse_html.gsub(/\x02([^\x03]*)\x03/) { "<span class=\"blb-match\">#{$1.strip}</span>" }
+      lines << "\n=== after placeholder restore ==="
+      lines << "  restored snippet: #{restored[0, 300].inspect}"
+
+      # Now compare with actual blueletterbible_html output
+      full_output = blueletterbible_html(html, url)
+      conc_match = full_output[/blb-match[^<]*<\/span>/]
+      lines << "\n=== blueletterbible_html output (blb-match check) ==="
+      lines << "  contains blb-match: #{full_output.include?("blb-match")}"
+      lines << "  blb-match context: #{conc_match.inspect}"
+      # Show the concordance section
+      conc_start = full_output.index("blb-heading") ? full_output.rindex("<h4", full_output.index("Concordance") || 0) : nil
+      if conc_start
+        lines << "  concordance html (first 500 chars): #{full_output[conc_start, 500].inspect}"
+      end
+
+      lines.join("\n")
+    end
+
+    get "/debug/fetch" do
+      url = params[:url].to_s.strip
+      halt 400, "missing ?url=" if url.empty?
+      html = fetch_external_page(url)
+      halt 502, "fetch failed" unless html
+      content_type :html
+      rendered = url.match?(/blueletterbible\.org\/lexicon\//i) ? blueletterbible_html(html, url) : page_html(html, url)
+      blb_css = "<style>.blb-table{width:100%;border-collapse:collapse;font-size:.85rem;margin-bottom:.6rem}.blb-table th,.blb-table td{padding:3px 7px;border:1px solid #ddd}.blb-th{text-align:left;font-weight:normal;background:#f5f0e4;color:#555;width:38%}.blb-right{text-align:right}.blb-nowrap{white-space:nowrap;vertical-align:top}.blb-match{color:#b33!important;font-weight:700!important;font-style:italic!important}.blb-heading{font-size:.82rem;font-weight:600;margin:.7rem 0 .25rem;color:#555;text-transform:uppercase;letter-spacing:.04em}.blb-usage{font-size:.85rem}.blb-usage ol{margin:.1rem 0 .1rem 1.3rem;padding:0;list-style-type:decimal}.blb-usage ol ol{list-style-type:lower-alpha}.blb-usage li{margin-bottom:.15rem}</style>"
+      "<html><head>#{blb_css}</head><body style='font-family:sans-serif;max-width:800px;margin:2rem auto;padding:0 1rem'>#{rendered}</body></html>"
+    end
+
     get "/download/*" do
       requested = params["splat"].first.to_s
       real_path = safe_path(requested)
@@ -680,8 +911,15 @@ module MarkdownServer
       html = fetch_external_page(url)
       halt 502, '{"error":"fetch failed"}' unless html
 
-      title = page_title(html).sub(/ [-–] .*/, "").strip
-      JSON.dump({ title: title, html: page_html(html, url) })
+      if url.match?(/blueletterbible\.org\/lexicon\//i)
+        raw = page_title(html)
+        title = raw.match(/^([GH]\d+ - \w+)/i)&.[](1)&.sub(" - ", " – ") ||
+                raw.sub(/ [-–] .*/, "").strip
+        JSON.dump({ title: title, html: blueletterbible_html(html, url) })
+      else
+        title = page_title(html).sub(/ [-–] .*/, "").strip
+        JSON.dump({ title: title, html: page_html(html, url) })
+      end
     end
 
     get "/search/?*" do

data/lib/markdown_server/version.rb

@@ -1,3 +1,3 @@
 module MarkdownServer
-  VERSION = "0.5.6"
+  VERSION = "0.5.8"
 end

data/views/layout.erb

@@ -909,6 +909,22 @@
     .link-preview-popup th, .link-preview-popup td { border: 1px solid #ddd; padding: 0.3rem 0.5rem; }
     .link-ctx-popup-body th, .link-preview-popup th { background: #f5f0e4; }
 
+    /* Blue Letter Bible popup tables */
+    .blb-table { width: 100%; border-collapse: collapse; font-size: 0.85rem; margin-bottom: 0.6rem; }
+    .blb-table th, .blb-table td { padding: 3px 7px; border: 1px solid #ddd; }
+    .blb-th { text-align: left; font-weight: normal; background: #f5f0e4; color: #555; width: 38%; }
+    .blb-right { text-align: right; }
+    .blb-nowrap { white-space: nowrap; vertical-align: top; }
+    .blb-match { color: #b33 !important; font-weight: 700 !important; font-style: italic !important; }
+    .blb-heading { font-size: 0.82rem; font-weight: 600; margin: 0.7rem 0 0.25rem; color: #555; text-transform: uppercase; letter-spacing: 0.04em; }
+    .blb-usage { font-size: 0.85rem; }
+    .blb-usage ol { margin: 0.1rem 0 0.1rem 1.3rem; padding: 0; list-style-type: decimal; }
+    .blb-usage ol ol { list-style-type: lower-alpha; }
+    .blb-usage ol ol ol { list-style-type: lower-roman; }
+    .blb-usage ol ol ol ol { list-style-type: lower-alpha; }
+    .blb-usage li { margin-bottom: 0.15rem; }
+    .blb-usage p { margin: 0; }
+
     /* Footnote tooltips */
     .footnote-tooltip {
       position: absolute;

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: markdownr
 version: !ruby/object:Gem::Version
-  version: 0.5.6
+  version: 0.5.8
 platform: ruby
 authors:
 - Brian Dunn