Six checks · Text or picture · What each proves
The six checks, one by one
A PDF can show a page while storing either characters, a picture, or both. Characters form a text layer that can often be selected, searched, copied, and counted. A page picture is an image of the page. It may look exactly like printed text, yet the letters are not stored as readable characters. The difference matters because appearance alone does not reveal what the file contains.
The six checks below move from direct signs to weaker clues. Selecting a line, searching for a visible word, and copying a sentence test whether characters can be reached. Enlarging a page tests appearance, not the underlying material. A count or an outside search can add context, but each has other explanations. The useful result is a pattern in which several checks point in the same direction.
This page is an independently written reference from pdf91.com, a newly built address with no business behind it. It names no file, document, product, or vendor, and it does not host, accept, convert, or publish documents. Its scope is narrower: explain what each ordinary reading check establishes, then mark the point where that signal stops being conclusive.
The checks that need nothing installed
The most direct checks use the document as it already appears. Place the pointer on a line and attempt to select part of it. If a selection follows individual words or a stretch of the line, the page carries accessible characters somewhere in its structure. A page that does not allow any selection may be picture-only, although a restriction can produce the same outward result.
A second direct route is to search within the document for a word that is plainly visible. A matching result shows that the searched word exists as characters that the document search can reach. No result is a useful indication of a page picture, but it is not final proof. The spelling may differ internally, the text layer may be incomplete, or access rules may prevent the search from seeing it.
Selecting and copying as a pair
Selection answers whether some text-like material can be reached, while copying tests what happens when that material leaves the page. Select a sentence and paste it into a plain text editor. If the sentence arrives in a readable order, the result supports the presence of usable characters. This is stronger than judging letter shapes because it observes the material carried out of the page.
The pasted result can also expose a mixed or damaged structure. Missing words, strange ordering, or unreadable symbols do not show that the page contains only pictures. They can indicate that the available text layer is incomplete or arranged differently from the visible page. Conversely, successful copying does not prove that every visible word is represented, since a picture may sit above a separate text layer.
| Try to select a line with the pointer | Where a selection follows the line, the page carries a text layer; where nothing selects at all, the page is likely a picture. |
|---|---|
| Search inside the document for a word you can see | A hit shows that the word exists as characters; no hit while the word is plainly visible points to a picture of a page. |
| Copy a sentence and paste it into a plain text editor | Text that arrives intact came from characters; text that arrives as nonsense, or not at all, points the other way. |
| Enlarge the page far beyond its normal size | Letters that stay crisp at any size may still be characters or a sharp image, so this check alone decides nothing. |
| Run a word count over the whole document | A very low count against a long document suggests pictures rather than text, though a short real document gives the same reading. |
| Look at what a search from outside the file returns | A document whose contents are never found by an outside search may hold no characters, but access restrictions produce the same result. |
Why enlarging a page settles the least
Enlarging the page is easy to do, but it examines visual sharpness rather than stored content. Letters that remain crisp at a large size may come from characters rendered by the document, or from a sharp page picture. Blurred edges can suggest an image, yet clarity cannot establish the opposite. The check therefore belongs near the beginning of inspection, not at the end of it.
A display can make a picture look clean enough to resemble real text, especially when the original page image is sharp. Rendering can also make characters appear less clear at certain sizes. Neither appearance settles whether selection, search, or copying will work. Enlargement is useful for noticing visual quality, but it supplies no independent answer about the text layer.
- Sharp letters can belong to characters or to a sharp picture.
- Blurred edges suggest image quality, not the absence of characters.
- A large view tests appearance rather than selectable document structure.
- Enlargement cannot confirm that every visible word is stored as text.
Searching for a word you can already see
Choose a plainly visible word and search for that same word inside the document. A hit means the word is present as searchable characters in the material the search can reach. It does not necessarily mean the visible letters came from those characters, because a picture can be displayed together with a text layer. The result is positive evidence, not a complete description of the page.
A visible word can be missed for several reasons that do not all point to pictures. The internal characters may be absent, incorrectly recognized, split in an unexpected way, or unavailable to the search. A protected document may also block access. Repeating the check with another clear word and comparing it with selection or copying gives the search result a firmer meaning.
Where the checks stop being conclusive
Every check in the table can be defeated. A picture of a page that also carries a text layer behind it will select and search while showing you the picture, and a genuine text document that is protected will refuse to let you select anything.
A word count is a rough signal at best. A real document can be short and a long picture-only document can still produce a count from its descriptive fields.
No single check settles the question on its own. Two checks that agree are worth more than one, and a reader who needs certainty should treat the result as a strong indication rather than a proof.
Nothing on this page tells a reader what to do with the answer. Whether the lack of a text layer matters depends on the use in hand, and that judgement stays with the reader.
The checks that are only a weak signal
A whole-document word count can help form an initial impression, especially when a long-looking document produces almost no words. It remains only a rough signal. A short document made from genuine characters can have a low count, while descriptive fields associated with a long picture-only document can produce some count. The number therefore needs a comparison with direct checks.
An outside search supplies another indirect clue. If the contents never appear there, the file may contain no characters that can be indexed. That result can also arise because access is restricted, the material is not available to the outside search, or the search does not reach that file. Absence from outside results cannot settle the structure of the document.
- A low word count can describe a short text document.
- Descriptive fields can create words around a picture-only document.
- Outside results can be absent because access is restricted.
- Indirect clues need support from selection, search, or copying.
Combining two checks into a reading
The clearest reading comes from agreement between checks that examine different behaviour. Selection followed by readable copying gives a stronger indication than enlargement alone. A visible word that can be found through document search and then copied in order supports the same reading. If direct checks disagree, the document may combine a page picture with a text layer, or a restriction may be interfering.
No single result should be treated as a final proof. A page picture can carry a hidden or accompanying text layer, allowing selection and search while the visible page remains an image. A genuine text document can refuse selection because of protection. The checks describe signals and limits; the decision about whether the result matters stays with the reader and the use in hand.
Frequently asked questions
What does selecting a line establish?
When a selection follows words or part of a line, the page carries accessible characters in some layer. That establishes a text signal, not that every visible letter is stored as text. A picture can appear above a text layer, and a restriction can block selection even where characters exist.
Why can a search miss a visible word?
The visible word may be part of a page picture rather than stored characters. It can also be absent from an incomplete text layer, represented differently internally, or hidden from the search by access restrictions. A missed word is therefore evidence worth comparing with selection and copying, not a complete diagnosis.
What does enlarging a page prove?
Enlargement proves only something about how the page looks at a larger view. Crisp letters may be rendered from characters or may belong to a sharp picture. Soft edges may suggest an image without proving that no text layer exists. Visual sharpness cannot settle the document's underlying structure.
When is a word count misleading?
A word count misleads when its size is treated as a direct measure of visible writing. A short genuine text document can have few words, while descriptive fields connected with a picture-only document can add words. The count is useful as a rough signal only when read beside more direct checks.
Does an outside search settle the question?
No. Failure to find the document's contents outside the file may suggest that searchable characters are absent, but restricted access can produce the same result. An outside search is indirect and cannot establish the file's internal structure. Selection, document search, and copying provide more relevant evidence.
This version of the reference was reviewed on September 20, 2026. Its purpose remains limited to reading signals in general: identify what a check shows, recognize its competing explanations, and avoid treating appearance or one isolated result as proof.
Where this record's facts originate
The page was written from scratch in September 2026. There was no material to carry over: the Wayback crawl index held no usable capture of the domain or any subdomain of it, and the search index returned no page the address had ever put out. A site-limited search surfaced unrelated pages belonging to other people, one of them a vehicle fault-code reference and one a Russian-language academic document collection, each mentioning the string inside its own unrelated text; that is an echo pointing at the address and was not treated as a source. The reference below describes the ordinary difference between stored characters and stored pictures of a page in general terms only: no product, vendor, standard number, version, file size or named document appears on it, and none has been made up. Where the behaviour of a file cannot be described without guessing at one program's conduct, the page says so and stops. Checked on September 20, 2026.