TROFF 到 pdf 或 docx 转换？

Question

我有部分完成的 TROFF 格式的博士论文（？） - 所以它可以从当时的 UNIX 站进行激光打印。它有一层 Atari 1stWord 封装，可能仍然位于标题处，但本质上是无关紧要的，因为它只是一个在家中使用的文本编辑器 - 远离大学的 Sun 工作站。

\n\n

谁能帮我将其转换为任何类型的所见即所得格式？pdf？文档？

\n\n

\xc2\xac066010303050000132002006006010000\n\xc2\xac1\xc2\xac\xc2\xac\xc2\xac1\n\xc2\xac2\xc2\xac#\xc2\xac\xc2\xac1\n\xc2\xacF0110000001\n\xc2\xac9[............................................................]0010\n \xc3\x87\n.\\" pic\n.\\" tbl\n.\\" eqn\n.\\" roff -ms\n.\\"\n.EQ\ndelim $$\nglobal\n.EN\n.ND\n.nr PO 1.5i\n.nr LL 6.0i\n.nr PS 9\n.nr VS 18\n.fp 1 H\n.TL\n.ps 14\nCHAPTER 7\n.AU\n.ps 14\nKNOWLEDGE ENGINEERING\n.PP\n.ps 9\n.NH\nIntroduction\n.PP\nIn the previous chapter we outlined a system which may be used to\nhandle uncertainty in a domain, using probabilities\non a causal network.\nIn this chapter we will to discuss how the knowledge within\na domain may be acquired and manipulated into the form\nof a probabilistic causal network - with specific reference to\nthe two domains which represent\nthe process of dating \ntimber-framed buildings, using\n(i) architectural/historical\nand\n(ii) dendrochronological evidence.\nOur ultimate goal here is to\nascertain, as best we may, the correct date of construction of\nvernacular buildings.\nAs mentioned in Chapter 2,\none of the main problems we face is the resolution of\nthe potential conflict between\ndendrochronological and archaeological predictions for the date of\na building.\nIf the two\nbodies of knowledge produce results which are totally\nin agreement, we are left with no other conclusion than that\nthe predicted date-range/period is the correct one,\nhowever there is rarely complete agreement and this disagreement\nmay be very difficult to resolve.\n.PP\nWe will derive networks on the two domains and give numerical\n \xc3\x87examples of propagation of evidence through these networks,\nclosing the chapter with a short exposition of\nhow we may use probabilistic methods in conjunction with\ncausal networks to resolve this mentioned conflict.\nThis use of probabilistic methods is contrary\nto widely adhered to\nnotions concerning the unsuitability of this approach - from\nthe general concerns over the epistemological adequacy of\nprobabilistic methods in artificial intelligence expressed by\nMcCarthy and Hayes (1969 p.490), to recent doubts about the implementational\nmechanics of probabilistic formalisms to be found in Graham and\nLlewelyn-Jones (1988 p.88 et seqq.).\n

\n

Answer 1

现在大多数 Linux 都带有GNU 版本的 Troff。

在我的 Linux Mint 系统上，该groff-base软件包已安装，但要通过 gropdf 输出实用程序生成 PDF，groff还需要安装该软件包。在 Solaris 上，您可能需要从源代码安装它，但有关此主题的手册相当稀疏。可以从这里下载源代码

安装后您可以运行：

groff -pet -ms -Tpdf in.troff > out.pdf

（假设您的输入位于in.troff），但您应该删除第一个单词的标题内容（在 之前.\" pic），否则您的输出如下所示：