add text wrapping

Adds a `--wrap_width <n>` for markdown formatting.

This required big changes to `output.rs`. It also inspired even more
changes: I should rework it to more of a stream model, where I just keep
track of the indentations, and treat newlines as special. But for now,
this works.

resolves #192

Cargo.toml

@@ -3,7 +3,7 @@ name = "mdq"
 version = "0.1.0"
 edition = "2021"
 license = "MIT OR Apache-2.0"
-description = "Select specific elements in a Markdown document" 
+description = "Select specific elements in a Markdown document"
 repository = "https://github.com/yshavit/mdq"
 
 # See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
@@ -15,6 +15,7 @@ paste = "1.0"
 regex = "1.10.4"
 serde = { version = "1", features = ["derive"] }
 serde_json = "1.0"
+termsize = "0.1.9"
 
 [dev-dependencies]
 indoc = "2"

src/cli.rs

@@ -1,6 +1,7 @@
 use crate::fmt_md::ReferencePlacement;
 use crate::link_transform::LinkTransform;
-use clap::{Parser, ValueEnum};
+use clap::error::ErrorKind;
+use clap::{CommandFactory, Parser, ValueEnum};
 use std::borrow::Cow;
 use std::fmt::{Display, Formatter};
 
@@ -28,6 +29,15 @@ pub struct Cli {
     #[arg(long, short, default_value_t = OutputFormat::Markdown)]
     pub(crate) output: OutputFormat,
 
+    /// The number of characters to wrap text at. This is only valid when the output format is
+    /// markdown.
+    ///
+    /// Certain elements (like section headings and link definitions) will never be wrapped, and the
+    /// wrapping will never break a word; it will only ever be along existing whitespace. In
+    /// particular, this means the wrapping will never add hyphens, and it will never break URLs.
+    #[arg(long)]
+    pub(crate) wrap_width: Option<usize>,
+
     #[arg(
         short = ' ',
         hide = true,
@@ -64,6 +74,25 @@ impl Cli {
             },
         }
     }
+
+    pub fn extra_validation(&self) -> bool {
+        match self.output {
+            OutputFormat::Json => {
+                if matches!(self.wrap_width, Some(_)) {
+                    let _ = Cli::command()
+                        .error(
+                            ErrorKind::ArgumentConflict,
+                            "Can't set text width with JSON output format",
+                        )
+                        .print();
+                    false
+                } else {
+                    true
+                }
+            }
+            OutputFormat::Markdown | OutputFormat::Md => true,
+        }
+    }
 }
 
 #[derive(Debug, Copy, Clone, PartialEq, Eq, PartialOrd, Ord, ValueEnum)]

src/fmt_md.rs

@@ -120,13 +120,15 @@ impl<'s, 'md> MdWriterState<'s, 'md> {
             }
             MdElemRef::Section(Section { depth, title, body }) => {
                 out.with_block(Block::Plain, |out| {
-                    for _ in 0..*depth {
-                        out.write_str("#");
-                    }
-                    if !title.is_empty() {
-                        out.write_str(" ");
-                        self.inlines_writer.write_line(out, title);
-                    }
+                    out.without_wrapping(|out| {
+                        for _ in 0..*depth {
+                            out.write_str("#");
+                        }
+                        if !title.is_empty() {
+                            out.write_str(" ");
+                            self.inlines_writer.write_line(out, title);
+                        }
+                    });
                 });
                 self.write_md(out, Self::doc_iter(body), false);
                 self.write_link_refs_as_needed(out);
@@ -136,7 +138,9 @@ impl<'s, 'md> MdWriterState<'s, 'md> {
             }
             MdElemRef::ThematicBreak => {
                 if !prev_was_thematic_break {
-                    out.with_block(Block::Plain, |out| out.write_str("   -----"));
+                    out.with_block(Block::Plain, |out| {
+                        out.without_wrapping(|out| out.write_str("   -----"));
+                    });
                 }
                 self.prev_was_thematic_break = true;
             }
@@ -440,9 +444,11 @@ impl<'s, 'md> MdWriterState<'s, 'md> {
                         LinkLabel::Text(identifier) => out.write_str(identifier.deref()),
                         LinkLabel::Inline(text) => self.inlines_writer.write_line(out, text),
                     }
-                    out.write_str("]: ");
-                    out.write_str(&link_def.url);
-                    self.inlines_writer.write_url_title(out, &link_def.title);
+                    out.without_wrapping(|out| {
+                        out.write_str("]: ");
+                        out.write_str(&link_def.url);
+                        self.inlines_writer.write_url_title(out, &link_def.title);
+                    });
                     newline(out);
                 }
             }
@@ -463,7 +469,7 @@ impl<'s, 'md> MdWriterState<'s, 'md> {
     }
 
     fn line_to_string(&mut self, line: &'md Line) -> String {
-        let mut out = Output::new(String::with_capacity(line.len() * 10)); // rough guess
+        let mut out = Output::without_text_wrapping(String::with_capacity(line.len() * 10)); // rough guess
         self.inlines_writer.write_line(&mut out, line);
         out.take_underlying().unwrap()
     }
@@ -2184,7 +2190,7 @@ pub mod tests {
         let (ctx, nodes) = inputs;
         nodes.iter().for_each(|n| VARIANTS_CHECKER.see(n));
 
-        let mut out = Output::new(String::default());
+        let mut out = Output::without_text_wrapping(String::default());
         write_md(options, &mut out, &ctx, nodes.into_iter());
         let actual = out.take_underlying().unwrap();
         assert_eq!(&actual, expect);