From bd97ecb9eb047e9fc58706616d70672668a38e08 Mon Sep 17 00:00:00 2001
From: Krzosa Karol <krzosa.karol@gmail.com>
Date: Fri, 12 Jul 2024 08:27:24 +0200
Subject: [PATCH] Big update

---
 src/transcript_browser/read_pdf.cpp | 76 +++++++++++++++++++++++++++++
 src/transcript_browser/read_srt.cpp | 63 ++++++++++++++++++++++++
 2 files changed, 139 insertions(+)
 create mode 100644 src/transcript_browser/read_pdf.cpp
 create mode 100644 src/transcript_browser/read_srt.cpp

diff --git a/src/transcript_browser/read_pdf.cpp b/src/transcript_browser/read_pdf.cpp
new file mode 100644
index 0000000..938bb21
--- /dev/null
+++ b/src/transcript_browser/read_pdf.cpp
@@ -0,0 +1,76 @@
+struct PDFPage {
+    String string;
+    int    number;
+};
+
+struct PDF {
+    String         filename;
+    Array<PDFPage> pages;
+};
+
+// @todo: pull request the object close
+// @todo: something needs to be done with unicode codepoints
+PDF pdfioReadPDF(Allocator allocator, String filename) {
+    Assert(filename.data[filename.len] == 0);
+    PDF result      = {};
+    result.pages    = {allocator};
+    result.filename = filename;
+
+    pdfio_file_t *file = pdfioFileOpen(filename.data, NULL, NULL, NULL, NULL);
+    Assert(file);
+    defer { pdfioFileClose(file); };
+
+    char   buffer[1024];
+    size_t page_count = pdfioFileGetNumPages(file);
+    result.pages.reserve(page_count);
+
+    for (size_t page_i = 0; page_i < page_count; page_i += 1) {
+        pdfio_obj_t *obj = pdfioFileGetPage(file, page_i);
+        if (obj == NULL) continue;
+        defer { pdfioObjClose(obj); };
+        PDFPage *page = result.pages.alloc();
+        page->number  = (int)page_i;
+
+        Scratch       scratch((Arena *)allocator.object);
+        Array<String> strings     = {scratch};
+        size_t        num_streams = pdfioPageGetNumStreams(obj);
+        for (size_t stream_i = 0; stream_i < num_streams; stream_i += 1) {
+            pdfio_stream_t *st = pdfioPageOpenStream(obj, stream_i, true);
+            if (st == NULL) continue;
+            defer { pdfioStreamClose(st); };
+
+            bool first = true;
+            while (pdfioStreamGetToken(st, buffer, sizeof(buffer))) {
+                if (buffer[0] == '(') {
+                    if (first) {
+                        first = false;
+                    }
+
+                    strings.add(Copy(scratch, buffer + 1));
+                } else if (!strcmp(buffer, "Td") || !strcmp(buffer, "TD") || !strcmp(buffer, "T*") || !strcmp(buffer, "\'") || !strcmp(buffer, "\"")) {
+                    first = true;
+                }
+            }
+        }
+        page->string = Merge(allocator, strings, "");
+    }
+
+    return result;
+}
+
+#if 1
+extern "C" void OutputDebugStringA(const char *);
+void            Printf(const char *string, ...) {
+    Scratch scratch;
+    STRING_FORMAT(scratch, string, result);
+    OutputDebugStringA(result.data);
+}
+#else
+    #define Printf(...) (0)
+#endif
+
+void PrintPDF(PDF pdf) {
+    For(pdf.pages) {
+        Printf("%d, %.*s\n", it.number, FmtString(it.string));
+    }
+}
\ No newline at end of file
diff --git a/src/transcript_browser/read_srt.cpp b/src/transcript_browser/read_srt.cpp
new file mode 100644
index 0000000..b198dec
--- /dev/null
+++ b/src/transcript_browser/read_srt.cpp
@@ -0,0 +1,63 @@
+struct TimeString {
+    uint16_t hour;
+    uint16_t minute;
+    uint16_t second;
+    String   string;
+};
+
+Array<TimeString> ParseSrtFile(Arena *arena, String filename) {
+    String        content = ReadFile(*arena, filename);
+    Array<String> lines   = Split(*arena, content, "\n");
+
+    IterRemove(lines) {
+        IterRemovePrepare(lines);
+        it = Trim(it);
+        if (it.len == 0) remove_item = true;
+    }
+
+    long              section_number = 1;
+    Array<TimeString> time_strings   = {*arena};
+    for (int i = 0; i < lines.len;) {
+        String it0 = lines[i++];
+        long   num = strtol(it0.data, NULL, 10);
+        Assert(section_number == num);
+        section_number += 1;
+
+        TimeString item = {};
+        String     it1  = lines[i++];
+        item.hour       = (uint16_t)strtol(it1.data, NULL, 10);
+        item.minute     = (uint16_t)strtol(it1.data + 3, NULL, 10);
+        item.second     = (uint16_t)strtol(it1.data + 6, NULL, 10);
+
+        String next_section_number = Format(*arena, "%d", section_number);
+        while (i < lines.len && lines[i] != next_section_number) {
+            String it   = lines[i];
+            item.string = lines[i];
+            time_strings.add(item);
+            i += 1;
+        }
+    }
+
+    IterRemove(time_strings) {
+        IterRemovePrepare(time_strings);
+        if (i > 0 && AreEqual(time_strings[i - 1].string, time_strings[i].string, true)) {
+            remove_item = true;
+        }
+    }
+
+    return time_strings;
+}
+
+String FindVideoForSRT(Array<String> &filenames, String srt_path) {
+    String base = ChopLastPeriod(srt_path); // .srt
+    base        = ChopLastPeriod(base);     // .en
+
+    For(filenames) {
+        if (StartsWith(it, base)) {
+            if (EndsWith(it, ".mkv") || EndsWith(it, ".webm") || EndsWith(it, ".mp4")) {
+                return it;
+            }
+        }
+    }
+    return {};
+}