filter control characters from content-level actual text

Seeing backspaces, which are maybe meant to represent underscores. Upsets XML, which just doesn't allow them.
pdf-raku · Oct 19, 2023 · f1b67e3 · f1b67e3
1 parent cd8a12b
commit f1b67e3
Show file tree

Hide file tree

Showing 10 changed files with 32 additions and 45 deletions.
diff --git a/.github/workflows/test.yml b/.github/workflows/test.yml
@@ -18,7 +18,7 @@ jobs:
           #- windows-latest
         raku-version:
           - 'latest'
-          - '2021.12'
+          - '2022.07'
     runs-on: ${{ matrix.os }}
     steps:
       - uses: actions/checkout@v2

diff --git a/README.md b/README.md
@@ -188,12 +188,10 @@ my PDF::XObject::Image $img .= open: "t/images/lightbulb.gif";
 my $figure = $doc.Figure: $gfx, $img, :position[50, 70], :Alt("A light-bulb");
 ```
 
-An [PDF::XObject::Form](https://pdf-raku.github.io/PDF-Class-raku/PDF/XObject/Form) may be associated with a marked content
-sub-tree. This is achieved by marking the form against a document fragment, then calling `do` to repeatably
-render the form, while inserting the fragment, as demonstrated below:
+A [PDF::XObject::Form](https://pdf-raku.github.io/PDF-Class-raku/PDF/XObject/Form) may be associated with a document fragment. The
+form can then be rendered, and the fragment inserted into the document, by repeatedly calling `do` on the fragment, as demonstrated below:
 
 ```raku
-
 use PDF::Tags;
 use PDF::Tags::Elem;
 use PDF::Class;
@@ -222,20 +220,11 @@ $page.graphics: -> $gfx {
         };
     }
 
-    # multiple rendering of the form, and insertion of its structure tree
+    # multiple rendering of the form, and insertion into the structure tree
     $doc.do($gfx, $form-frag, :position[150, 70]);
     $doc.do($gfx, $form-frag, :position[150, 20]);
 }
-
 ```
-
-To insert an XObject Form that has marked content:
-
-1. Create a new fragment element.
-2. Create the Form XObject, marking content against the fragment
-3. The `do` method can then be used to both render and insert
-a copy of the fragment into the structure tree.
-
 ### Links
 
 Links are usually contained in a block element, such as a `Paragraph`. If

diff --git a/docs/PDF/Tags/Mark.md b/docs/PDF/Tags/Mark.md
@@ -92,7 +92,7 @@ The Marked Content ID within the content stream. These are usually numbered in s
 
     method value() returns PDF::Content::Tag
 
-The low-level [PDF::Content::Tag](https://pdf-raku.github.io/PDF-Content-raku) object, which contains further details on the tag:
+The low-level [PDF::Content::Tag](https://pdf-raku.github.io/PDF-Content-raku/PDF/Content/Tag) object, which contains further details on the tag:
 
   * `canvas` - The owner of the content stream; a PDF::Page or PDF::XObject::Form object.
 

diff --git a/docs/PDF/Tags/Node.md b/docs/PDF/Tags/Node.md
@@ -13,7 +13,7 @@ Methods
 
 ### method cos
 
-Returns the underlying [PDF::Class](https://pdf-raku.github.io/PDF-Class-raku) or [PDF::Content](https://pdf-raku.github.io/PDF-Content-raku) object. The [PDF::Tags::Node](https://pdf-raku.github.io/PDF-Tags-raku/PDF/Tags/Node) subclass and [PDF::COS](https://pdf-raku.github.io/PDF-raku) type are mapped as follows:
+Returns the underlying [PDF::Class](https://pdf-raku.github.io/PDF-Class-raku) or [PDF::Content](https://pdf-raku.github.io/PDF-Content-raku/PDF/Content) object. The [PDF::Tags::Node](https://pdf-raku.github.io/PDF-Tags-raku/PDF/Tags/Node) subclass and [PDF::COS](https://pdf-raku.github.io/PDF-raku) type are mapped as follows:
 
 <table class="pod-table">
 <thead><tr>

diff --git a/docs/PDF/Tags/XML-Writer.md b/docs/PDF/Tags/XML-Writer.md
@@ -12,10 +12,10 @@ Synopsis
 --------
 
     use PDF::Class;
-    use PDF::Tags;
+    use PDF::Tags::Reader;
     use PDF::Tags::XML-Writer;
     my PDF::Class $pdf .= open: "t/write-tags.pdf";
-    my PDF::Tags $tags .= read: :$pdf;
+    my PDF::Tags::Reader $tags .= read: :$pdf;
     my PDF::Tags::XML-Writer $xml-writer .= new: :debug, :root-tag<Docs>;
     # atomic write
     say $xml-writer.Str($tags);

diff --git a/docs/index.md b/docs/index.md
@@ -188,12 +188,10 @@ my PDF::XObject::Image $img .= open: "t/images/lightbulb.gif";
 my $figure = $doc.Figure: $gfx, $img, :position[50, 70], :Alt("A light-bulb");
 ```
 
-An [PDF::XObject::Form](https://pdf-raku.github.io/PDF-Class-raku/PDF/XObject/Form) may be associated with a marked content
-sub-tree. This is achieved by marking the form against a document fragment, then calling `do` to repeatably
-render the form, while inserting the fragment, as demonstrated below:
+A [PDF::XObject::Form](https://pdf-raku.github.io/PDF-Class-raku/PDF/XObject/Form) may be associated with a document fragment. The
+form can then be rendered, and the fragment inserted into the document, by repeatedly calling `do` on the fragment, as demonstrated below:
 
 ```raku
-
 use PDF::Tags;
 use PDF::Tags::Elem;
 use PDF::Class;
@@ -222,20 +220,11 @@ $page.graphics: -> $gfx {
         };
     }
 
-    # multiple rendering of the form, and insertion of its structure tree
+    # multiple rendering of the form, and insertion into the structure tree
     $doc.do($gfx, $form-frag, :position[150, 70]);
     $doc.do($gfx, $form-frag, :position[150, 20]);
 }
-
 ```
-
-To insert an XObject Form that has marked content:
-
-1. Create a new fragment element.
-2. Create the Form XObject, marking content against the fragment
-3. The `do` method can then be used to both render and insert
-a copy of the fragment into the structure tree.
-
 ### Links
 
 Links are usually contained in a block element, such as a `Paragraph`. If

diff --git a/lib/PDF/Tags/Mark.rakumod b/lib/PDF/Tags/Mark.rakumod
@@ -84,13 +84,23 @@ class PDF::Tags::Mark
         fail "todo: update marked content attributes";
         callsame();
     }
+
+    sub sanitize(Str $_) {
+        # actual text sometimes have backspaces, etc?
+        .subst(
+            /<[ \x0..\x8 ]>/,
+            '',
+            :g
+        );
+    }
+
     method ActualText {
         $.attributes unless $!atts-built;
-        $!actual-text //= PDF::COS::TextString.COERCE: $_
+        $!actual-text //= sanitize PDF::COS::TextString.COERCE: $_
             with %!attributes<ActualText>;
         $!actual-text;
     }
-    method remove-actual-text {
+    method remove-actual-text is DEPRECATED {
         with $.ActualText {
             $!actual-text = Nil;
             $!value.attributes<ActualText>:delete;

diff --git a/lib/PDF/Tags/Node.rakumod b/lib/PDF/Tags/Node.rakumod
@@ -76,7 +76,7 @@ class PDF::Tags::Node {
 Returns the underlying L<PDF::Class> or L<PDF::Content> object. The L<PDF::Tags::Node> subclass and L<PDF::COS> type are mapped as follows:
 
 =begin table
-PDF::Tags::Node object | PDF::Class object |Base class | Notes
+PDF::Tags::Node object | PDF::Class object  |Base class | Notes
 =================================================
 PDF::Tags | PDF::StructTreeRoot | PDF::Tags::Node::Parent | PDF structure tree root
 PDF::Tags::Elem | PDF::StructElem | PDF::Tags::Node::Parent | Intermediate structure element node

diff --git a/lib/PDF/Tags/Node/Parent.rakumod b/lib/PDF/Tags/Node/Parent.rakumod
@@ -16,7 +16,7 @@ class PDF::Tags::Node::Parent
     has UInt $!elems;
     has      $.style is rw; # Computed CSS style
 
-    method elems is also<Numeric> {
+    method elems(::?CLASS:D:) is also<Numeric> {
         $!elems //= do with $.cos.kids {
             when Hash { 1 }
             default { .elems }

diff --git a/lib/PDF/Tags/XML-Writer.rakumod b/lib/PDF/Tags/XML-Writer.rakumod
@@ -49,16 +49,15 @@ method !chunk(Str $s is copy, UInt $depth = 0) {
 method !line(|c) { $!feed = True; self!chunk(|c); $!feed = True; }
 method !frag(|c) { $*inline ?? self!chunk(|c) !! self!line(|c) }
 
-sub html-escape(Str $_) {
+sub xml-escape(Str $_) {
     .trans:
         /\&/ => '&amp;',
         /\</ => '&lt;',
         /\>/ => '&gt;',
-
 }
 multi sub str-escape(@a) { @a.map(&str-escape).join: ' '; }
 multi sub str-escape(Str $_) {
-    html-escape($_).trans: /\"/ => '&quote;';
+    xml-escape($_).trans: /\"/ => '&quote;';
 }
 multi sub str-escape(Pair $_) { str-escape(.value) }
 multi sub str-escape($_) is default { str-escape(.Str) }
@@ -233,7 +232,7 @@ multi method stream-xml(PDF::Tags::Elem $node, UInt :$depth is copy = 0) {
                 }
             }
             
-            given html-escape($_) {
+            given xml-escape($_) {
                 my $frag = do {
                     when $omit-tag.so { $_ }
                     when .so { '<%s%s>%s</%s>'.sprintf($name, $att, $_, $name) }
@@ -287,7 +286,7 @@ multi method stream-xml(PDF::Tags::Mark $node, :$depth!) {
 
 multi method stream-xml(PDF::Tags::Text $_, :$depth!) {
     if .Str -> $text {
-        self!chunk(html-escape($text), $depth);
+        self!chunk(xml-escape($text), $depth);
     }
 }
 
@@ -297,7 +296,7 @@ method !marked-content(PDF::Tags::Mark $node, :$depth!) {
             when PDF::Tags::Mark {
                 my $text = self!marked-content($_, :$depth);
             }
-            when PDF::Tags::Text { html-escape(.Str) }
+            when PDF::Tags::Text { xml-escape(.Str) }
             default { die "unhandled tagged content: {.WHAT.raku}"; }
         }
         @text.join;
@@ -319,10 +318,10 @@ method !marked-content(PDF::Tags::Mark $node, :$depth!) {
 =head2 Synopsis
 
     use PDF::Class;
-    use PDF::Tags;
+    use PDF::Tags::Reader;
     use PDF::Tags::XML-Writer;
     my PDF::Class $pdf .= open: "t/write-tags.pdf";
-    my PDF::Tags $tags .= read: :$pdf;
+    my PDF::Tags::Reader $tags .= read: :$pdf;
     my PDF::Tags::XML-Writer $xml-writer .= new: :debug, :root-tag<Docs>;
     # atomic write
     say $xml-writer.Str($tags);