[UX] Make search more obvious in PDF viewer
Categories
(Firefox :: PDF Viewer, enhancement, P3)
Tracking
()
People
(Reporter: phlsa, Unassigned)
References
Details
(Whiteboard: [pdfjs-ux] [ux] p=8)
Updated•11 years ago
|
Updated•11 years ago
|
Updated•11 years ago
|
Updated•11 years ago
|
Reporter | ||
Updated•11 years ago
|
Updated•11 years ago
|
Updated•10 years ago
|
To demonstrate the problem described in this bug:
FIRSTLY, point Firefox to the Mozilla PDF.js "online demo" at https://mozilla.github.io/pdf.js/web/viewer.html
There is a spyglass near the upper left, between the icons for "Toggle Sidebar" and "Previous Page".
Using this spyglass searches the entire PDF document. For example, searching for the text "the" results in 985 matches.
The Esc key removes the spyglass drop-down. Cool.
Ctrl-F also activates the PDF.js spyglass (and does not open the otherwise standard Firefox search bar). Cool.
SECONDLY, in another tab or window, make sure Firefox's PDF.js is enabled, and point Firefox directly to the same document again at
"blob:https://mozilla.github.io/d6a647e3-6436-400f-9e74-a875411eb09f" (the "blob:" prefix is essential!).
This time, there is no spyglass for the PDF.js integrated text search.
Nor does Ctrl-F invoke the PDF.js integrated text search. Instead, Ctrl-F invokes the standard Firefox search bar, which however yields incorrect results: searching for the text "the" never results in the correct 985 matches, but shows some smaller number, depending on what part of, and how much of, the document is currently visible.
I invite anyone to take a few minutes to carry out the above two experiments, and to see which version they prefer.
Quite aside from personal taste and from any philosophical preferences, the fact remains that the standard Firefox text search in a PDF.js document (second experiment) yields clearly incorrect and inconsistent numbers of matches. This problem was reported in various Firefox bugs over the years, but seemingly never fixed.
It would therefore seem appropriate to:
EITHER: finally fix the standard Firefox text search to display the correct number of matches in PDF.js documents.
OR: abandon the standard Firefox text search for PDF.js documents, and re-instate the PDF.js integrated text search spyglass, so that PDF.js works just the same in Firefox as it does in the PDF.js "online demo" described above (which presumably is how Mozilla intends PDF.js to be implemented?).
Please excuse my ignorance - I am only a user, not a developer - but what has to happen to get this bug promoted from "NEW" to "ASSIGNED"?
(Given that this bug describes an easily reproducable error, namely that Ctrl-F yields incorrect numbers of matches in PDF.js documents: see previous comment.)
Updated•3 years ago
|
Updated•3 years ago
|
Updated•3 years ago
|
Description
•