CN107102877A - A kind of adaptive approach of browser-cross plug-in unit - Google Patents

Abstract

The invention discloses a cross-browser plug-in self-adaptive method, which includes a standard browser plug-in, and the standard browser plug-in is self-adaptive to the browser to read the webpage files of the website, and specifically includes the following steps: step 1.1. The plug-in is loaded into the current browser; Step 1.2 The standard browser plug-in accesses the current browser kernel and reads the information of the current browser; Step 1.3 The standard browser plug-in then analyzes the web page file read by the browser to determine the web page Whether the file is compatible with the current browser; the step 1.4 standard browser plug-in is adjusted according to the type of the current browser kernel to the consistency of the parsed webpage file; the step 1.5 browser renders and displays the adjusted webpage file, and the present invention passes the standard The browser plug-in analyzes and adjusts the consistency of the webpage files, so that the webpage files can be rendered normally, which is conducive to saving costs and does not affect the customer's operation.

Description

An Adaptive Method for Cross-Browser Plug-ins

technical field

The invention relates to a cross-browser plug-in, in particular to a cross-browser standard plug-in, in particular to a cross-browser plug-in adaptive method.

Background technique

A browser refers to a software that can display the content of an HTML file (an application of the standard universal markup language) on a web server or file system, and allows users to interact with these files. In order to improve the scalability and ease of use of the browser, the browser needs to provide plug-ins to support users to access rich multimedia information on the Internet.

The browser kernel is divided into Trident, Gecko, WebKit, and Presto. Trident (also known as MSHTML) is a typesetting engine developed by Microsoft, and its representative product is Internet Explorer, so it is also called the IE core. Gecko is a set of open source web page typesetting engine written in C++. The representative work of Gecko kernel is Mozilla Firefox. WebKit is an open source project, including some components from the KDE project and Apple. It is characterized by a clear source code structure and extremely fast rendering speed. The disadvantage is that the compatibility with webpage codes is not high, which causes some non-standard webpages to not be displayed normally. The main representative works are Safari and Google's browser Chrome. Presto is a browser typesetting engine developed by Opera Software, which supports web page re-layout following DOM and Script syntax events. The main representative work is Opera.

Browser plug-in technology is mainly used to expand browser functions, including registering MIME types, opening its own working window in the browser serial port, accepting and processing keyboard and mouse practices, using URL to obtain and transmit data from the network, and adding pointers to new URL hotkeys or hotzones, control HTML pages, etc. Due to the diversity of browser kernels, different kernels support plug-ins of different standards.

Because plug-ins are developed and released by different publishers, and their technical levels vary, plug-in programs are likely to conflict with other running programs, resulting in various page errors, runtime errors, etc., blocking normal browsing, Moreover, different browser development standards are inconsistent, resulting in browser plug-ins that can only run in specific browsers. Different plug-ins need to be released for different browsers, and plug-ins cannot be unified. At the same time, when a browser reads a web page file, When the webpage file has a format or other errors that cannot be properly recognized by the current browser, it will not be displayed normally.

Contents of the invention

The present invention provides a cross-browser plug-in self-adaptive method, which is used to solve the problem existing in the prior art: when the browser reads a webpage file, when the webpage file has a format or is wrongly written, the browser cannot render normally The problem of display, in order to solve the above-mentioned technical problems and achieve the technical effect of modifying the errors of the webpage file and being able to be displayed normally by the current browser without replacing the webpage file, the present invention is realized through the following technical solutions: a The adaptive method of the cross-browser plug-in framework includes a standard browser plug-in, and the standard browser plug-in adapts to the browser to read the web page file of the website, specifically including the following steps:

Step 1.1 loads the standard browser plug-in into the current browser;

Step 1.2 The standard browser plug-in accesses the current browser kernel and reads the information of the current browser;

Step 1.3 The standard browser plug-in parses the webpage file read by the browser to determine whether the webpage file is compatible with the current browser;

Step 1.4 The standard browser plug-in adjusts the consistency of the parsed webpage file according to the type of the current browser kernel;

Step 1.5 The browser renders and displays the adjusted webpage file.

Workflow description: Because plug-ins are developed and released by different publishers, and their technical levels vary, plug-in programs are likely to conflict with other running programs, resulting in various page errors, runtime errors, etc., blocking normal browsing, and the development standards of different browsers are inconsistent, resulting in browser plug-ins that can only run in specific browsers, different plug-ins need to be released for different browsers, and it is impossible to achieve a unified plug-in. The specific development process is to first define a unified standard development platform. Standard browser plug-ins are developed through standard development platforms. First, standard browser plug-ins are created on the standard development platform, and finally standard browser plug-ins are created. The plug-in is sent to the browser, and the standard development platform is equipped with a standard framework of a parser and a class loader. A plug-in structure is set in the standard frame, and the plug-in structure includes at least one root directory. At least four subdirectories, namely general assembly, plug-in files, library files, and general configuration. The general assembly contains common programs commonly used by standard browser plug-ins, and the general configuration contains common configurations and libraries commonly used by standard browser plug-ins. There are library files commonly used by standard browser plug-ins in the file, and there are multiple plug-in function modules under the plug-in files in the subdirectory, and each function module contains 4 units, which are assembly and script file , webpage files, and library files. The function of the assembly is to define the properties of the plug-in, such as name, version, type and other information. The web page file is to implement the function of the plug-in. Write the corresponding program code in 4 units, specifically refers to writing under the web page file, so as to form the corresponding plug-in function, the class loader of the standard framework loads and calls the plug-in file, general assembly and library file according to the general configuration , combined into corresponding standard browser plug-ins.

The standard browser plug-in can also be tested on the standard development platform, and there are alarms during the trial run. The standard browser plug-in developed on the standard development platform simulates the standard browser plug-in according to the virtual network environment. If there is a problem, such as the web page cannot be displayed normally, the picture cannot be opened normally, there will be a prompt message at this time, and the prompt message will be processed, and the standard browser plug-in will be reflected in the specific position of the corresponding plug-in structure, and the standard The webpage files of the browser plug-in are modified to meet the needs of various browser kernels.

Organizing, packaging and sending the standard browser plug-ins that have been successfully tested or returned to the modified test run successfully, and sent to the browser, this process is the specific process of loading the standard browser plug-ins into the current browser.

The development of standard browser plug-ins is conducive to unification. When plug-ins need to be developed, corresponding programs are written under the plug-in structure of the standard development platform, and then assembled to form plug-ins with corresponding functions, which can reduce the waste of human resources and at the same time support different browsers. Types are also fully adaptable.

The standard browser plug-in developed on the standard development platform can also be tested here. If the operation is successful, it will enter the next step to store the browser, access, etc. When the trial run is unsuccessful, the standard browser plug-in will be returned to the plug-in structure , make the wrong modification, the required requirement has been achieved.

The standard browser plug-in can access the current browser kernel and read the information of the current browser. It is worth noting that the access process and the process of reading information are common knowledge in the art, so too much elaboration will not be given here.

The rendering engine of the current browser displays the requested content on the window. The content is that the browser can read the webpage files of the website. Usually, the rendering engine can display html, xml and other documents and pictures, but it cannot display special files. Normal rendering, while standard browser plug-ins can render special files and then display them on the browser window, such as PDF files, CEB files, Flash Player and other special types of files.

For special files that cannot be displayed normally, the standard browser plug-in can analyze the webpage files at this time, and through parsing, the webpage files that can be recognized by the current browser are reached, and the corresponding webpage files are adjusted consistently, so that Enables the current browser to render and display normally.

In order to better realize that the standard browser plug-in is parsing the webpage file, and achieves the completeness of the analysis, it is further defined, the specific steps for the standard browser plug-in in the step 1.3 to analyze the webpage file that the browser reads the website are as follows: :

Step 1.3.1: the standard browser plug-in performs grammatical analysis on the source file of the web page file;

Step 1.3.2: The standard browser plug-in builds a parse tree for the source file of the web page file;

Step 1.3.3: The standard browser plug-in converts the parse tree into a document that can be recognized by the current browser according to the type of the current browser kernel.

The specific process of parsing a webpage file by a standard browser plug-in. Parsing a webpage file is to convert the file into a meaningful structure, that is, a structure that allows the code to be understood and used. A webpage file is usually a code format. Standard browsing The browser plug-in first performs grammatical analysis on the code of the web page file. The code format has certain grammatical rules. Grammar refers to the grammatical rules based on the basis of the document, that is, the language or format of the document. The format of each file that can be parsed must have A specific grammar consisting of vocabulary and grammatical rules. The standard browser plug-in contains a parser, and the parser is generally divided into a syntax analyzer and a lexical analyzer. It is worth noting that the syntax analyzer and the lexical analyzer are well known in the art, so I will not elaborate too much here. The lexical analyzer is mainly lexical analysis, which decomposes the input into symbols. The symbol is the vocabulary of the language, which is equivalent to the words in the dictionary we are familiar with. The grammatical analyzer mainly analyzes the grammar. The grammatical analysis refers to the application of grammar to the language. The parser generally works in two parts, the lexical analyzer is responsible for decomposing the input into legal symbols, the grammatical analyzer analyzes the document according to the grammatical rules of the language, and the parse tree finally converts the document according to the above lexical analysis and grammatical analysis Build as a parse tree.

The resulting parse tree is not the final structure, and the parser needs to convert the parse tree into a document that can be recognized by the current browser kernel.

This process describes the parsing process in more detail, which makes the solution of the present invention more perfect. By parsing the webpage files, the webpage files that cannot be rendered are converted into files that can be recognized by the current browser.

For documents that can be identified by the conversion bit, because the developer's development level is uneven, there are many human errors. In order to better realize the present invention, it is further defined. In the step 1.4, according to the type of the current browser kernel Adjust the consistency of the parsed webpage files. The specific steps for adjusting the consistency are:

Step 1.4.1: Recognize the symbol of the document and pass it to the builder;

Step 1.4.2: The builder reconstructs the received symbols into a build tree that can adapt to the current browser kernel;

Step 1.4.3: The builder modifies the symbols of the re-established construction tree according to the current browser kernel;

Step 1.4.4: Convert the modified build tree to a web page file.

Recognize the symbols of the document through the parser. Symbol recognition refers to parsing the input into symbols according to the algorithm of symbol recognition, and passing it to the tree builder, and reading the next character to recognize the next symbol until After processing all the inputs, symbol recognition is mainly to identify the start tag, end tag, attribute name and attribute value.

The tree construction phase will be built, elements will be appended to the tree, and each node recognized by the symbol recognizer will be processed by the tree builder. The builder re-establishes the construction tree according to the current browser kernel, and also modifies the symbols during the construction process, which has reached the effect of normal rendering.

In order to realize the present invention more concretely, and to realize better self-adaptation, it is further defined that in the step 1.4.3, the builder modifies the re-established construction tree including wrong or irregular compilation of the source files in the webpage files Make repairs. Modify the errors in the webpage files, mainly to modify the non-standard writing methods and redundant symbols, so as to achieve the effect of normal rendering.

In order to implement the present invention more specifically, it is further defined that the repair is adding or reducing symbols of the source file.

In order to save programs and save time, after the webpage file is modified, it is further defined that between the steps 1.4.3 and 1.4.4, a trial run of the modified webpage file is also included. If the trial run is successful, it can be rendered normally. If it is not successful, it will go back to modify until the modification is successful.

In order to make the present invention more perfect during operation, it is further defined that an alarm prompt is also provided during the trial operation. When there is an unsuccessful trial operation, an alarm will be prompted immediately, and the alarm information will be displayed, so that the webpage file can be modified faster and more accurately.

Compared with the prior art, the present invention has the following advantages and beneficial effects:

(1) The present invention uses a standard browser plug-in to analyze and adjust the consistency of webpage files, so that webpage files can be rendered normally, which is beneficial to saving costs and does not affect customer operations;

(2) The present invention adjusts the consistency of webpage files, especially for special files, such as PDF files, so that the product has stronger versatility and wider application range;

(3) The present invention can reduce errors and improve the efficiency of use by performing trial operation after consistency adjustment;

(4) In the present invention, by setting an alarm prompt during the trial operation, the alarm prompt can often alarm and process error information, deal with corresponding errors in time, and save time.

Description of drawings

Other characteristics, objects and advantages of the present invention will become more apparent by reading the detailed description of non-limiting embodiments with reference to the following drawings:

Fig. 1 is the flowchart of cross-browser plug-in adaptive method of the present invention;

Fig. 2 is the flow chart that cross-browser plug-in of the present invention parses webpage file;

Fig. 3 is the flowchart of the consistency adjustment after the web page file is parsed by the cross-browser plug-in of the present invention;

Fig. 4 is a flow chart of the standard browser plug-in of the present invention;

Fig. 5 is the plug-in framework schematic diagram of standard browser plug-in of the present invention;

Fig. 6 is a browser component diagram of the present invention.

detailed description

Embodiments of the present invention are described in detail below, examples of which are shown in the drawings, wherein the same or similar reference numerals designate the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the figures are exemplary only for explaining the present invention and should not be construed as limiting the present invention.

Example 1:

The main flow of this embodiment is shown in Figure 1 to Figure 6: 1. A cross-browser plug-in adaptive method includes a standard browser plug-in, and the standard browser plug-in adapts to the browser to read the web page file of the website, specifically including the following steps:

Step 1.1 loads the standard browser plug-in into the current browser;

Step 1.2 The standard browser plug-in accesses the current browser kernel and reads the information of the current browser;

Step 1.3 The standard browser plug-in parses the webpage file read by the browser to determine whether the webpage file is compatible with the current browser;

Step 1.4 The standard browser plug-in adjusts the consistency of the parsed webpage file according to the type of the current browser kernel;

Step 1.5 The browser renders and displays the adjusted webpage file.

Specific implementation method: For webpage projects developed based on different rendering engines, their formats and grammars are different, and many websites are not designed and developed in accordance with W3C standards, resulting in rendering engines and standard browser plug-ins when parsing webpage files. , cannot be displayed properly. In this case, it is necessary to adjust the parsed webpage file, improve and adjust the format of the parsed webpage file, and realize the self-adaptation of the standard browser plug-in.

The standard browser plug-in is developed according to the standard development platform. First, the standard development platform needs to be defined, and the standard browser plug-in is set on the standard development platform. The standard development platform has a parser and a class loader. The standard development platform standard The plug-in structure is set in the framework. The plug-in structure includes at least one root directory. Under the root directory, there are subdirectories of general assembly, plug-in files, library files, and general configuration. Among them, only commonly used programs are stored in the general assembly file. The common configuration is the basic configuration of the plug-in, and the library file is the commonly used library file. There is physical isolation between each subdirectory. Different plug-in function modules are set according to different functions under the plug-in file, and each plug-in function module also has physical isolation at the same time. Set the plug-in structure and put the plug-in corresponding function For classification, each plug-in function module is a function. The plug-in function module contains all the files required to realize the function, such as assembly, script file, page file and picture. Write the corresponding program under this plug-in function module, and then it can be realized. corresponding function.

The plug-in function module is derived from the basic architecture model of the plug-in. Generally, a complete plug-in is composed of 4 parts, which are assembly, script file, library file and web page file. Usually, the role of the library file is to define the plug-in. Attributes, such as name, version, type, etc.; while the webpage file page can contain pictures, links, and even non-text elements such as music and programs, which are the functions of plug-ins. The directory of the plug-in is organized and stored according to the directory of the development standard platform.

After the standard browser plug-in is developed, each function will be encapsulated in the plug-in function module. There are more source file directories stored in the plug-in function module. Each standard browser plug-in has corresponding functions and corresponding interfaces. , business logic, data and other codes, because each plug-in function module is also physically isolated, developers can develop their own functions, and the class loader parses these codes, and independently tests, deploys and Upgrades, once developed, are assembled on the development standard platform by classloaders. The class loader on the standard development platform parses and runs the standard browser plug-in. According to the standard framework of the standard development platform, there are more plug-in modules in the standard framework.

After the development of the standard browser plug-in is completed, a trial run will be required. After the trial run is successful, it will be packaged and sent to the browser.

The standard browser plug-in first accesses the current browser type and reads the corresponding information. When the current browser reads the web page file of the website and cannot display it, the standard browser plug-in analyzes the web page and conforms to the standard. The specific analysis and Adjust the following examples to illustrate with the HTML format of the web page file,

For the client HTML file, when the standard browser plug-in obtains the file, it needs to adjust the consistency of the file. The specific parsing process is as follows: parsing is divided into two sub-processes: grammar analysis and lexical analysis. Grammar analysis is to decompose the input into symbols. Symbols are the vocabulary of the language. Grammar analysis refers to the rules applied to the language. Lexical analysis is responsible for decomposing the input. For legal symbols, analyze the document structure of HTML according to the grammar rules of the language, so as to build a parse tree.

After obtaining the parse tree, the standard browser plug-in converts the parse tree into a web document that can be recognized by the current browser, and converts it into a web document that can be recognized in time, but there is a possibility that the web document is written incorrectly or in the wrong format. Adjust the consistency of web documents, the specific consistency adjustment process is as follows:

1) Symbol recognition

HTML symbols include start tags, end tags, attribute names, and attribute values. Symbol recognition means that according to the symbol recognition algorithm, the input is parsed into symbols and passed to the tree builder, and the next character is read to recognize the next symbol until all inputs are processed.

The symbol recognition algorithm is relatively complicated, and this principle is explained with an HTML page:

<html>

<body>

hello world

</body>

</html>

The initial state is "Data State". When a "<" character is encountered, the state changes to "Tag open state". Reading a character from a to z will generate a start tag symbol, and the state changes to "Tag name state" accordingly. Keep this state until ">" is read, and each character is appended to the symbol name. In the example, an html symbol is created. When ">" is read, the current symbol is completed. At this time, the state returns to "Data state", and "<body>" repeats this process. At this point, both html and body tags are recognized. Now, going back to the "Data state", reading the character "H" in "Hello world" will create and recognize a glyph, here a glyph will be generated for each character in "Hello world". Do this until you encounter the "<" in "</body>". Now, back to the "Tag open state", reading the next character "/" creates a closed tag symbol, and the state transitions to the "Tag name state", which remains until a ">" is encountered. Then, generate a new label symbol and go back to "Data state". The following "</html>" will be treated the same as "</body>".

2) Rebuild the build tree

In the tree construction phase, the DOM tree with Document as the root will be modified, and elements will be appended to the tree. Each node generated by the symbol recognizer will be processed by the tree constructor. Each symbol is defined in the specification. The corresponding Dom element, the corresponding Dom element will be created. In addition to being added to the DOM tree, these elements will also be added to the open element stack. This stack is used to correct nested unmatched and unclosed tags. This algorithm is also described by a state machine, and all states use insert mode.

Continue to explain the principle of building a tree:

This stage takes as input the sequence of symbols produced by the symbol recognition stage and starts building. The first is "initialmode", which will convert to "before html" mode after receiving the html symbol, in which the symbol is reprocessed. At this point, an HTMLHtmlElement element is created and attached to the root Document object. The state changes to "before head" at this point, and when the body symbol is received, even if there is no head symbol here, an HTMLHeadElement element will be automatically created and appended to the tree. Now, go to "in head" mode, then "after head". At this point, the body symbol will be processed again, an HTMLBodyElement will be created and inserted into the tree, and at the same time, transferred to "in body" mode. Then, the character notation for the string "Hello world" is received, the first character will cause a text node to be created and inserted, and the other characters will be appended to it. When receiving the end of body symbol, transfer to "afterbody" mode, and then receive the end of html symbol, this symbol means transfer to "after after body" mode, when the end of file is received, the entire parsing process ends.

3) Adaptive adjustment

Since some HTML pages are written according to the Gecko standard, some Windows users are written according to the Trident standard, and some are written according to the Webkit standard, the format of identification symbols between different rendering engines is inconsistent, The plug-in self-adaptation analyzes the input file according to the kernel of the current browser, automatically modifies and improves the elements, so that the file can be displayed normally. E.g:

<table>

<table>

<tr><td>inner table</td></tr>

</table>

<tr><td>outer table</td></tr>

</table>

The current browser kernel is webkit, and the input file can be rendered normally after analyzing and processing the input file through the adaptive plug-in. If there is no adaptive plug-in, webkit will turn the nested table into two sibling tables:

<table>

<tr><td>outer table</td></tr>

</table>

<table>

<tr><td>inner table</td></tr>

</table>

On the other hand, for some irregularly written and wrong HTML pages, the adaptive plug-in can automatically repair tags and elements.

When the client is a CSS file, when the adaptive plug-in obtains the file, it needs to adjust the consistency of the file. The specific settings are:

Unlike html, css is a context-free grammar and can be defined in a format similar to BNF. A rule set has one or an optional number of selectors separated by commas and spaces (S means space). Each rule set consists of braces and one or more semicolon-separated declarations within the braces. Declarations and selectors are defined later.

For example, browsers with a webkit core use Flex and Bison parser generators to automatically generate parsers from CSS grammar files. Bison creates a bottom-up parser, Firefox uses a top-down parser. They all parse each css file into a style sheet object, each object contains css rules, css rule objects contain selectors and declaration objects, and some other objects that conform to css syntax. The adaptive plug-in is adjusted according to the different results generated by the parser to realize the self-adaptation of CSS.

The trial run is carried out after the webpage file is adaptively adjusted. This trial run is beneficial to check errors and facilitate modification. At the same time, an alarm prompt is set during the trial run. When encountering a problem, even the prompt is more efficient.

It can be understood that the structure of the parser according to an embodiment of the present invention, such as the working principle and working process of components such as a lexical analyzer and a syntax analyzer, are all prior art and well known to those skilled in the art. Here No more detailed description will be given.

Although the embodiments of the present invention have been shown and described, those skilled in the art can understand that various changes, modifications, substitutions and modifications can be made to these embodiments without departing from the principle and spirit of the present invention. The scope is defined by the claims and their equivalents.

Claims (7)

1. an adaptive method of cross-browser plug-in, comprising standard browser plug-in, it is characterized in that: described standard browser plug-in is to the webpage file adaptive adjustment of browser reading website, specifically comprises the following steps: Step 1.1 loads the standard browser plug-in into the current browser; Step 1.2 The standard browser plug-in accesses the current browser kernel and reads the information of the current browser; Step 1.3 The standard browser plug-in parses the webpage file read by the browser to determine whether the webpage file is compatible with the current browser; Step 1.4 The standard browser plug-in adjusts the consistency of the parsed webpage file according to the type of the current browser kernel; Step 1.5 The browser renders and displays the adjusted webpage file. 2. the self-adaptive method of a kind of cross-browser plug-in according to claim 1, is characterized in that: standard browser plug-in in described step 1.3 the concrete steps that the webpage file that browser reads website is analyzed again is: Step 1.3.1: the standard browser plug-in performs grammatical analysis on the source file of the web page file; Step 1.3.2: After grammar analysis, the standard browser plug-in builds a parse tree for the source file of the web page file; Step 1.3.3: The standard browser plug-in converts the parse tree into a document that can be recognized by the current browser according to the type of the current browser kernel. 3. the self-adaptive method of a kind of cross-browser plug-in according to claim 2, it is characterized in that: in described step 1.4, carry out consistency adjustment to the webpage file of parsing again according to the type of current browser kernel, concrete consistency The adjustment steps are: Step 1.4.1: Recognize the symbols of the document that can be recognized by the current browser and pass it to the builder of the standard browser plug-in; Step 1.4.2: the builder reconstructs the received symbols into a construction tree that can adapt to the current browser kernel; Step 1.4.3: The builder modifies the symbols of the re-established construction tree according to the current browser kernel; Step 1.4.4: Convert the modified construction tree into a web page file that can be rendered by the current browser. 4. The adaptive method of a kind of cross-browser plug-in according to claim 3, it is characterized in that: in the described step 1.4.3, builder revises the re-established construction tree and includes the error of the source file in the webpage file Or non-standard writing to fix. 5. A cross-browser plug-in adaptive method according to claim 4, characterized in that: said repair is to increase or decrease the symbol of the source file. 6. the self-adaptive method of a kind of cross-browser plug-in according to claim 3, is characterized in that: between described step 1.4.3 and step 1.4.4, also comprise that the webpage file after modification is carried out test run . 7. A cross-browser plug-in adaptive method according to claim 6, characterized in that: an alarm prompt is also set during the trial run.