dany/pdfio
1
0
Fork 0
mirror of https://github.com/michaelrsweet/pdfio.git synced 2025-07-19 15:29:46 +02:00
Code Issues Packages Projects Releases Wiki Activity

Update pdfioinfo example to support Acrobat Form dictionaries as well as indirect references (Issue #114)

Browse Source
This commit is contained in:
Michael R Sweet
2025-04-04 21:24:42 -04:00
parent 0bd9edc845
commit 130cef8702
5 changed files with 311 additions and 47 deletions
Download Patch File Download Diff File Expand all files Collapse all files

123
doc/pdfio.html
View File

@ -1,13 +1,13 @@
<!DOCTYPE html>
<html lang="en-US">
<head>
<title>PDFio Programming Manual v1.5.0</title>
<title>PDFio Programming Manual v1.5.2</title>
<meta http-equiv="Content-Type" content="text/html;charset=utf-8">
<meta name="generator" content="codedoc v3.8">
<meta name="author" content="Michael R Sweet">
<meta name="language" content="en-US">
<meta name="copyright" content="Copyright © 2021-2025 by Michael R Sweet">
<meta name="version" content="1.5.0">
<meta name="version" content="1.5.2">
<style type="text/css"><!--
body {
background: white;
@ -251,7 +251,7 @@ span.string {
<body>
<div class="header">
<p><img class="title" src="pdfio-512.png"></p>
<h1 class="title">PDFio Programming Manual v1.5.0</h1>
<h1 class="title">PDFio Programming Manual v1.5.2</h1>
<p>Michael R Sweet</p>
<p>Copyright © 2021-2025 by Michael R Sweet</p>
</div>
@ -1165,11 +1165,26 @@ main(<span class="reserved">int</span> argc, <span clas
{
<span class="reserved">const</span> <span class="reserved">char</span> *filename; <span class="comment">// PDF filename</span>
pdfio_file_t *pdf; <span class="comment">// PDF file</span>
<span class="reserved">const</span> <span class="reserved">char</span> *author; <span class="comment">// Author name</span>
time_t creation_date; <span class="comment">// Creation date</span>
<span class="reserved">struct</span> tm *creation_tm; <span class="comment">// Creation date/time information</span>
<span class="reserved">char</span> creation_text[<span class="number">256</span>]; <span class="comment">// Creation date/time as a string</span>
<span class="reserved">const</span> <span class="reserved">char</span> *title; <span class="comment">// Title</span>
pdfio_dict_t *catalog; <span class="comment">// Catalog dictionary</span>
<span class="reserved">const</span> <span class="reserved">char</span> *author, <span class="comment">// Author name</span>
*creator, <span class="comment">// Creator name</span>
*producer, <span class="comment">// Producer name</span>
*title; <span class="comment">// Title</span>
time_t creation_date, <span class="comment">// Creation date</span>
modification_date; <span class="comment">// Modification date</span>
<span class="reserved">struct</span> tm *creation_tm, <span class="comment">// Creation date/time information</span>
*modification_tm; <span class="comment">// Modification date/time information</span>
<span class="reserved">char</span> creation_text[<span class="number">256</span>], <span class="comment">// Creation date/time as a string</span>
modification_text[<span class="number">256</span>], <span class="comment">// Modification date/time human fmt string</span>
range_text[<span class="number">255</span>]; <span class="comment">// Page range text</span>
size_t num_pages; <span class="comment">// PDF number of pages</span>
<span class="reserved">bool</span> has_acroform; <span class="comment">// Does the file have an AcroForm?</span>
pdfio_obj_t *page; <span class="comment">// Object</span>
pdfio_dict_t *page_dict; <span class="comment">// Object dictionary</span>
size_t cur, <span class="comment">// Current page index</span>
prev; <span class="comment">// Previous page index</span>
pdfio_rect_t cur_box, <span class="comment">// Current MediaBox</span>
prev_box; <span class="comment">// Previous MediaBox</span>
<span class="comment">// Get the filename from the command-line...</span>
@ -1182,14 +1197,20 @@ main(<span class="reserved">int</span> argc, <span clas
filename = argv[<span class="number">1</span>];
<span class="comment">// Open the PDF file with the default callbacks...</span>
pdf = pdfioFileOpen(filename, <span class="comment">/*password_cb*/</span>NULL, <span class="comment">/*password_cbdata*/</span>NULL,
<span class="comment">/*error_cb*/</span>NULL, <span class="comment">/*error_cbdata*/</span>NULL);
pdf = pdfioFileOpen(filename, <span class="comment">/*password_cb*/</span>NULL,
<span class="comment">/*password_cbdata*/</span>NULL, <span class="comment">/*error_cb*/</span>NULL,
<span class="comment">/*error_cbdata*/</span>NULL);
<span class="reserved">if</span> (pdf == NULL)
<span class="reserved">return</span> (<span class="number">1</span>);
<span class="comment">// Get the title and author...</span>
author = pdfioFileGetAuthor(pdf);
title = pdfioFileGetTitle(pdf);
<span class="comment">// Get the title, author, etc...</span>
catalog = pdfioFileGetCatalog(pdf);
author = pdfioFileGetAuthor(pdf);
creator = pdfioFileGetCreator(pdf);
has_acroform = pdfioDictGetType(catalog, <span class="string">&quot;AcroForm&quot;</span>) != PDFIO_VALTYPE_NONE;
num_pages = pdfioFileGetNumPages(pdf);
producer = pdfioFileGetProducer(pdf);
title = pdfioFileGetTitle(pdf);
<span class="comment">// Get the creation date and convert to a string...</span>
<span class="reserved">if</span> ((creation_date = pdfioFileGetCreationDate(pdf)) &gt; <span class="number">0</span>)
@ -1202,12 +1223,76 @@ main(<span class="reserved">int</span> argc, <span clas
snprintf(creation_text, <span class="reserved">sizeof</span>(creation_text), <span class="string">&quot;-- not set --&quot;</span>);
}
<span class="comment">// Get the modification date and convert to a string...</span>
<span class="reserved">if</span> ((modification_date = pdfioFileGetModificationDate(pdf)) &gt; <span class="number">0</span>)
{
modification_tm = localtime(&amp;modification_date);
strftime(modification_text, <span class="reserved">sizeof</span>(modification_text), <span class="string">&quot;%c&quot;</span>, modification_tm);
}
<span class="reserved">else</span>
{
snprintf(modification_text, <span class="reserved">sizeof</span>(modification_text), <span class="string">&quot;-- not set --&quot;</span>);
}
<span class="comment">// Print file information to stdout...</span>
printf(<span class="string">&quot;%s:\n&quot;</span>, filename);
printf(<span class="string">&quot; Title: %s\n&quot;</span>, title ? title : <span class="string">&quot;-- not set --&quot;</span>);
printf(<span class="string">&quot; Author: %s\n&quot;</span>, author ? author : <span class="string">&quot;-- not set --&quot;</span>);
printf(<span class="string">&quot; Created On: %s\n&quot;</span>, creation_text);
printf(<span class="string">&quot; Number Pages: %u\n&quot;</span>, (<span class="reserved">unsigned</span>)pdfioFileGetNumPages(pdf));
printf(<span class="string">&quot; Title: %s\n&quot;</span>, title ? title : <span class="string">&quot;-- not set --&quot;</span>);
printf(<span class="string">&quot; Author: %s\n&quot;</span>, author ? author : <span class="string">&quot;-- not set --&quot;</span>);
printf(<span class="string">&quot; Creator: %s\n&quot;</span>, creator ? creator : <span class="string">&quot;-- not set --&quot;</span>);
printf(<span class="string">&quot; Producer: %s\n&quot;</span>, producer ? producer : <span class="string">&quot;-- not set --&quot;</span>);
printf(<span class="string">&quot; Created On: %s\n&quot;</span>, creation_text);
printf(<span class="string">&quot; Modified On: %s\n&quot;</span>, modification_text);
printf(<span class="string">&quot; Version: %s\n&quot;</span>, pdfioFileGetVersion(pdf));
printf(<span class="string">&quot; AcroForm: %s\n&quot;</span>, has_acroform ? <span class="string">&quot;Yes&quot;</span> : <span class="string">&quot;No&quot;</span>);
printf(<span class="string">&quot; Number of Pages: %u\n&quot;</span>, (<span class="reserved">unsigned</span>)num_pages);
<span class="comment">// Report the MediaBox for all of the pages</span>
prev_box.x1 = prev_box.x2 = prev_box.y1 = prev_box.y2 = <span class="number">0.0</span>;
<span class="reserved">for</span> (cur = <span class="number">0</span>, prev = <span class="number">0</span>; cur &lt; num_pages; cur ++)
{
<span class="comment">// Find the MediaBox for this page in the page tree...</span>
<span class="reserved">for</span> (page = pdfioFileGetPage(pdf, cur);
page != NULL;
page = pdfioDictGetObj(page_dict, <span class="string">&quot;Parent&quot;</span>))
{
cur_box.x1 = cur_box.x2 = cur_box.y1 = cur_box.y2 = <span class="number">0.0</span>;
page_dict = pdfioObjGetDict(page);
<span class="reserved">if</span> (pdfioDictGetRect(page_dict, <span class="string">&quot;MediaBox&quot;</span>, &amp;cur_box))
<span class="reserved">break</span>;
}
<span class="comment">// If this MediaBox is different from the previous one, show the range of</span>
<span class="comment">// pages that have that size...</span>
<span class="reserved">if</span> (cur == <span class="number">0</span> ||
fabs(cur_box.x1 - prev_box.x1) &gt; <span class="number">0.01</span> ||
fabs(cur_box.y1 - prev_box.y1) &gt; <span class="number">0.01</span> ||
fabs(cur_box.x2 - prev_box.x2) &gt; <span class="number">0.01</span> ||
fabs(cur_box.y2 - prev_box.y2) &gt; <span class="number">0.01</span>)
{
<span class="reserved">if</span> (cur &gt; prev)
{
snprintf(range_text, <span class="reserved">sizeof</span>(range_text), <span class="string">&quot;Pages %u-%u&quot;</span>,
(<span class="reserved">unsigned</span>)(prev + <span class="number">1</span>), (<span class="reserved">unsigned</span>)cur);
printf(<span class="string">&quot;%16s: [%g %g %g %g]\n&quot;</span>, range_text,
prev_box.x1, prev_box.y1, prev_box.x2, prev_box.y2);
}
<span class="comment">// Start a new series of pages with the new size...</span>
prev = cur;
prev_box = cur_box;
}
}
<span class="comment">// Show the last range as needed...</span>
<span class="reserved">if</span> (cur &gt; prev)
{
snprintf(range_text, <span class="reserved">sizeof</span>(range_text), <span class="string">&quot;Pages %u-%u&quot;</span>,
(<span class="reserved">unsigned</span>)(prev + <span class="number">1</span>), (<span class="reserved">unsigned</span>)cur);
printf(<span class="string">&quot;%16s: [%g %g %g %g]\n&quot;</span>, range_text,
prev_box.x1, prev_box.y1, prev_box.x2, prev_box.y2);
}
<span class="comment">// Close the PDF file...</span>
pdfioFileClose(pdf);
@ -5081,6 +5166,10 @@ ssize_t pdfioStreamPeek(<a href="#pdfio_stream_t">pdfio_stream_t</a> *st, <span
</tbody></table>
<h4 class="returnvalue">Return Value</h4>
<p class="description"><code>true</code> on success, <code>false</code> on failure</p>
<h4 class="discussion">Discussion</h4>
<p class="discussion">This function writes a formatted string to a stream. In addition to the
standard <code>printf</code> format characters, you can use &quot;%N&quot; to format a PDF name
value (&quot;/Name&quot;) and &quot;%S&quot; to format a PDF string (&quot;(String)&quot;) value.</p>
<h3 class="function"><a id="pdfioStreamPutChar">pdfioStreamPutChar</a></h3>
<p class="description">Write a single character to a stream.</p>
<p class="code">