Skip to content

armgit / Filee

Public Imported from GitHub
0

Free, open-source file converter. Convert 263 formats offline by drag & drop — images, PDF, Word, Excel, PowerPoint, HWP/HWPX, video, audio, e-books and archives. No upload, no Office needed.

ENGINES.md: names in the code page choice, one ? per emoji, table cells, the DWG code page index

cre8ive-marketing0edb87bd5841
diff --git a/docs/ENGINES.md b/docs/ENGINES.md
index f9c105f1..366fab74 100644
--- a/docs/ENGINES.md
+++ b/docs/ENGINES.md
@@ -862,7 +862,10 @@ and writes the files; Filee's own renderer draws them with SkiaSharp (one drawin
which AutoCAD 2007 opens; kept from a 2007 source, as 2010. Before 2007, text is stored in the drawing's code page
(`$DWGCODEPAGE`), so `CadCodePages` picks the one that holds the most of the drawing's text (its own on a tie;
Korean for Hangul, Japanese for Kana, Big5 before GBK for Chinese, East Asian pages only for East Asian text) and
- writes any character it can't hold as `\U+XXXX`. Korean is written as `ANSI_949`, as AutoCAD names it (ACadSharp
+ writes any character it can't hold as `\U+XXXX` (one `?` for a character outside the BMP). Names count and are
+ escaped too: layers, blocks, text and dimension styles, linetypes, layouts and attribute tags, so a Korean layer in a
+ drawing with only English text gets `ANSI_949`. Text inside table cells (2004 and later) isn't escaped. In a DWG, the
+ header's code page index is ACadSharp's (`KSC5601` / `GB2312`), the DXF says `ANSI_949`; both read back the same. Korean is written as `ANSI_949`, as AutoCAD names it (ACadSharp
needs `KCS5601` to encode it, so the DXF header is renamed after writing). Objects the chosen version doesn't
have are left out with a note: tables and arc dimensions before 2004, multileaders and PDF underlays before 2007,
meshes before 2010. ACadSharp's DWG writer is younger than its reader