A PDF's body reads the text layer of the page it shows

The body of a PDF pane read as empty. It now reads MuPDF's text extraction of the page shown, the text search and n/N already use, read-only (a write is refused, as round 22's body refusal made it).

Co-Authored-By: Claude Opus 5.5 <[email protected]>

2 files changed, 37 insertions, 1 deletions

diff --git a/src/ninep/ctl.zig b/src/ninep/ctl.zig
index db058ff2..be90bb6f 100644
--- a/src/ninep/ctl.zig
+++ b/src/ninep/ctl.zig
@@ -2115,6 +2115,27 @@ test "a language server question that finds nothing, cannot be answered, or is a
     try testing.expectEqual(E.INVAL, wr(p, Node.of(img.serial, .body), "text\n").errno());
 }
 
+test "a PDF's body reads its page's text layer, and takes no write" {
+    if (comptime !pardes.pdf_enabled) return;
+    const p = try withFile(testing.allocator, "x\n");
+    defer p.deinit();
+    var tmp = testing.tmpDir(.{});
+    defer tmp.cleanup();
+    const bytes = try @import("mupdf").makeLinksTestPdf(testing.allocator);
+    defer testing.allocator.free(bytes);
+    try tmp.dir.writeFile(testing.io, .{ .sub_path = "doc.pdf", .data = bytes });
+    var dir_buf: [4096]u8 = undefined;
+    const dir = dir_buf[0..try tmp.dir.realPath(testing.io, &dir_buf)];
+    var line: [4200]u8 = undefined;
+    _ = wr(p, root_look, try std.fmt.bufPrint(&line, "{s}/doc.pdf\n", .{dir}));
+    p.sync();
+    const doc = p.panes[p.active].?;
+    try testing.expect(doc.hasPdf());
+    const body = Node.of(doc.serial, .body);
+    try testing.expect(std.mem.indexOf(u8, rd(p, body, 0, 4096).bytes, "target.txt") != null);
+    try testing.expectEqual(E.INVAL, wr(p, body, "more\n").errno());
+}
+
 test "every EINVAL a write gets says why, in its err record too; DEL is a control character in a line" {
     const p = try withFile(testing.allocator, "x\n");
     defer p.deinit();
diff --git a/src/ninep/pane.zig b/src/ninep/pane.zig
index 9e832df6..311a2429 100644
--- a/src/ninep/pane.zig
+++ b/src/ninep/pane.zig
@@ -257,7 +257,7 @@ pub fn fileSize(p: *Pardes, id: usize, f: PaneFile) u64 {
     const pf = &pane.fs;
     return switch (f) {
         // A terminal's body reads its screen and history, as text.
-        .body => if (pane.file == null and pane.isTerminal()) panes.Terminal.screenTextLen(pane) else bodyOf(pane).len,
+        .body => if (pane.file == null and pane.isTerminal()) panes.Terminal.screenTextLen(pane) else if (pdfText(p, pane)) |t| t.len else bodyOf(pane).len,
         .data, .xdata => bodyOf(pane).len,
         .tag => tagOf(p, pane).len,
         .name => nameOf(pane).len + 1,
@@ -367,6 +367,14 @@ fn readFlag(p: *Pardes, req: Req, on: bool) Reply {
     return tree.stagedReply(p, req);
 }
 
+/// The text layer of the page a PDF pane shows (MuPDF's extraction, kept
+/// with the page), null for any other pane.
+fn pdfText(p: *Pardes, pane: *Pane) ?[]const u8 {
+    if (comptime !pardes.pdf_enabled) return null;
+    const pv = &(pane.pdf orelse return null);
+    return pv.ensureText(p.pdf_gpa);
+}
+
 fn readBody(p: *Pardes, req: Req, id: usize, pane: *Pane) Reply {
     if (pane.file != null) {
         const text = bodyOf(pane);
@@ -379,6 +387,13 @@ fn readBody(p: *Pardes, req: Req, id: usize, pane: *Pane) Reply {
             .len = clip(n),
         } } };
     }
+    // A PDF's body is its text layer, the page shown's: read-only.
+    if (pdfText(p, pane)) |text| {
+        const off = @min(req.off, text.len);
+        const out = p.fs.stage(p.gpa);
+        out.appendSlice(p.gpa, text[off..][0..@min(text.len - off, req.size)]) catch return Reply.fail(req.tag, E.NOMEM);
+        return tree.stagedReply(p, req);
+    }
     if (req.handle != 0 and pane.isTerminal()) return screen.readSnapshot(p, req, pane);
     const text = panes.Terminal.screenTextAlloc(pane, p.gpa) catch
         return Reply.fail(req.tag, E.NOMEM);