↑ Return to SP80 Spinning

SP87 HTML into Spinning Text

Page no: SP87

Example / Test Evidences

Overview HTML into Spinning Text and Back

Spinning software destroys the HTML code.

Spinning software only accepts normal text.

Transfer HTML into normal text.

All HTML tags and syntax must be kept on output from spinning.


Transfer HTML into Spinning Input Text
  1. Transfer HTML into actual text and symbols that is ignored by spinning engine
    (ignored means not spinned)
  2. Create a mapping table between Symbols and the actual HTML tags
    –> Transfer Text Software does this
Transfer Spinning Output into HTML

3. Obtain the spinned text output.

4. Replace the symbols with HTML tags again (reverse mapping table).

Mapping Table Identifiers to HTML

We introduce a unique technique for keeping words or codes, which we do not want to be spinned. This technique is introduced for all needed HTML tags.

Full list of tags which use this technique:

  • All image tags, including their alt tags or title tags
  • A HREF links
  • Div, paragraphs
  • All classes and ids


Differences between code, symbol, identifier.
Identifier is unique.


Example Mapping Table
(created by the software)




Replace tags with identifiers

We create a function which detects which are all the problem tags (WHAT IS THIS ?)  and make a list with them.

When we put the tag into the list, then we replace it with a unique code. We make different list for different type of tags we have.

The solution will be the same as we use for the images. We will detect these words in the original text, no matter if we want to be capital-first-letter words or user-defined. We will replace them with a unique code . Example “Germany” will be “#1#”, which will means that this is the first replaced word.

When we get the spinned article, we will did the reverse replace. We will replace the unique code with the original word. “#1#” will be replaced with Germany.

This is a technology which is used for HTML tags and images.

More information can be found here (where)


Unique Identifier Generation

The identifiers must be unique

Avoid wrong replacement.

The syntax for generating the codes is simple. We have different code for different tag. The general syntax is “#(letter)(number)(number)”.

The first letter is the first letter of the tag we will search. The number is which number we firstly found this tag in the original text.



Example: “#I01#” = Image, first one,

Please put full HTML syntax from the mapping


#1# <src …>
#2# <  s

“#D02#” – Div, second one


Replace codes with tags on spinned article

When the article is spinned and we got it from the spinner provider, we go trough all of the text and search again. But this time not for the HTML tags, but for the unique tags (what is this?) we have. When we find a unique tag, we change it with the original  html content. Then we save the article.



Transfer Spinning Output into HTML


Testing of Text Transfer

Procedure of texting and fixing bugs:

1) Do a bug testing

1.1) Test all params: date, authors, categories and combine them.

1.2) See if spinning works correctly with all params

What does mean “spinning works”?


  • Text Transfer HTML –> Spinning Input
  • Spinning Output –> HTML

1.3) During test params, test all text changes before and after sending to the spinner

1.4) Make screenshots of the output to be sure everything is fine





Special Cases

Not necessarily part of Prototype.

We have problem with images from Blogspot.

They have strange sytaxis of the url of the image. This breaks our search & replace function and the tag was not replaced with a code.

Fixed –> Not what I want to see

Example: Want to see that it is fixed. Testing Evidences and Examples
Class and ID names

We had a problem with classes and IDs when there are more than one and there is space between them

Their syntaxis somehow brakes our search & replace function

Tested and fixed –> Not what I

 Example: Want to see that it is fixed. Testing Evidences and Examples




See more for Sa8x Spinning