markdownr 0.5.7 → 0.5.8

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: e9a74b1e181f0fd5361c73e64a6e00659451cf075e1e41ec4f45c0cee9167e9d
-  data.tar.gz: 44b5d07121159a9efadac6ee85bd4528d2490ec419b0752dbe15e1336e94eea2
+  metadata.gz: 44c223ecfa0f81c35d2d3d89a5c316eb2adc7f0a343db241185dc828268d3527
+  data.tar.gz: 7b87a1fed84d540c89bc92bea8de754eee3d680ba93f952ad4b23c7150aa06b3
 SHA512:
-  metadata.gz: 866416e07423170e5ec3c84e60da99682bfabb07b09cf1eb74f29e7e74f509c9982ec292641721ec0d054f5361bd6012fc00a34755ab46987c4462e63569926e
-  data.tar.gz: 1a18c15a57094d2a5d72670d760d82d1eeca80ab1c694cc550a7ff69e352c6dd9e889840e515b0a40c9e59a9b48e9aa85744d6983f37f33c572b90cdd39c28a8
+  metadata.gz: 8eb259d062a13af47d3126ea380777e01aba1f1e6bc9401fb2f05c8277e9a827d6e16f9299e96768236574114875c0f4d2b5ab273094513c71670afa765f3e12
+  data.tar.gz: 56fda7d165fecbec5b46d8a496d257aa38164684779225bfac6e1d9da00df3fb8da3e90fbd4c58c435f2680a21869271f06c4b3791df48b9ac1b6d64e7359277

data/lib/markdown_server/app.rb

@@ -672,23 +672,35 @@ module MarkdownServer
           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) ?
-              %(<span class="blb-match">#{h(word)}</span>) : h(word)
+              "\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;/, "&").gsub!(/&lt;/, "<").gsub!(/&gt;/, ">")
+          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)
@@ -749,6 +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)

data/lib/markdown_server/version.rb

@@ -1,3 +1,3 @@
 module MarkdownServer
-  VERSION = "0.5.7"
+  VERSION = "0.5.8"
 end

data/views/layout.erb

@@ -915,7 +915,7 @@
     .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; font-weight: 600; }
+    .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; }

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: markdownr
 version: !ruby/object:Gem::Version
-  version: 0.5.7
+  version: 0.5.8
 platform: ruby
 authors:
 - Brian Dunn