mupdf加载PDF显示中文乱码

现象

加载PDF显示乱码,提示非嵌入字体

 non-embedded font using identity encoding

调式

在pdf-font.c中加载字体

调试源码发现pdf文档的字体名字居然是GBK，估计又是哪个windows下写的pdf生成工具生成pdf

字体方法：

static pdf_font_desc *
load_cid_font(fz_context *ctx, pdf_document *doc, pdf_obj *dict, pdf_obj *encoding, pdf_obj *to_unicode)

具体加载

 pdf_to_name(ctx, pdf_dict_get(ctx, dict, PDF_NAME(BaseFont)));

解决

转码发现是宋体，简单判断转换下

  unsigned char utf8SimSun[]={0xe5,0xae,0x8b,0xe4,0xbd,0x93};unsigned char gbkSimSun[]={0xcb,0xce,0xcc,0xe5};

     if(0 == memcmp(basefont,utf8SimSun,sizeof(utf8SimSun))){fz_warn(ctx, "basefont:宋体UTF8");basefont = "SimSun";}if(0 == memcmp(basefont,gbkSimSun,sizeof(gbkSimSun))){fz_warn(ctx, "basefont:宋体GBK");basefont = "SimSun";}

PDF字体查看

可以借助福昕软件查看