1285323 九、發明說明： 【發明所屬之技術領域】 本發明技術是屬於電腦數據處理領域，尤指可供上網 瀏覽中文文件之電腦數據處理系統與方法。 【先前技術】 網際網路瀏覽器被廣泛地應用於搜尋及顯示電子文 件，這些電子文件係以HTML格式文件儲存於連結到網際 網路上之電腦伺服器裡，電腦使用者藉由上網就可以擷取 到這些HTML格式文件。 當一電腦使用者上網時，會透過瀏覽器之軟體來瀏覽 一些存放在網頁伺服器之網頁文件。使用者可以輸入一網 頁名稱（統稱為URL)來指定一網站以及其上之一文件檔， 讓瀏覽器去搜尋。 當一以HTML檔案表現之網頁文件被擷取後，瀏覽器 顯示該文件以提供使用者去閱覽。瀏覽器軟體係透過解譯 HTML檔來選文字檔裏的字型資料，再根據字型資料來顯示 文件裏的文字。瀏覽器所顯示之文字係依照在HTML文字 檔所設定之字體及字型資料逐字勾描其組成筆劃或外形 所形成；若設定之字體無法被瀏覽器認知，則瀏覽器會選 一預定之系統字體來替代，再用該預存之系統字體來顯示 該文件。 HTML語言定義了一些標示以及它們的屬性，讓一網頁 2 1285323 設計者能用它們來規範文件文句之内涵以呈現該網頁。然 而HTML語言定義於字體的標示僅限於系統所提供者，如 果網頁文件含有非系統字體則瀏覽器僅能用一系統預定 . 的字體來取代（這種替代方法描述於美國專利 • 5, 893, 915)，這種限制對顯示中文網頁文件會有變醜（不 • 會看到原貌）甚至變形，缺字（有些文字看不到）的現象 發生。 中文字可以用象形符號表示，其外形可以被自由表 • 達，例如圖一所示，其中第一個字’’台’’係由二個字根所形 成，二個字根可以用二種不同字體來組合該字；第二個字，， 灣，，係屬於雕刻體；第三個字’’心，，的二個筆劃用，，桃形，，來表 - 示，這三個字的字體皆不能被傳統瀏覽器所認知，使得被 . 瀏覽到的文件原貌走樣及變形。 職是之故，在網路上瀏覽中文網頁文件的技藝上，急 需一種可以提供可瀏覽中文網頁文件原貌之系統與方法。 【發明内容】 總結本發明之要項如下： 本發明提供一種方法及系統讓上網之電腦使用杳閱 覽到具有多字體之網頁的原貌文件。網頁提供者所採用多 子體包含儲存於系統裡或者不存在於系統字體裡I，這些 字體文字係先前技術之方法及系統所無法閱覽到的，在先 前技藝裡這些字體係由系統所指定之一個字體來取代，而 本發明技藝是可以閱覽到系統或非系統字體文字之文件 3 1285323 原貌的方法。 本發明系統採用一網頁字型伺服器連結到網頁伺服 器來重建原存於網頁伺服器之網頁文件，使之成為網頁字 型HTML檔，得以讓本發明所提之網頁字型瀏覽器來認知 網頁字型HTML檔並顯示出來文件原貌。本發明之網頁字 型伺服器除了本身提供系統字體字型資料外，並透過一公 共數據網路連結到一部外字伺服器來提供非系統字體字 型資料。這樣安排的系統可以讓第一次被接收到網頁字型 伺服器之網頁文件得以從網頁字型伺服器直接找網頁字 型或者間接從外字字型伺服器找出所要之網頁字型將文 件重製成網頁字型文件，再將網頁字型文件傳回網頁伺服 器以取代原第一次被接收的網頁文件，使得第二次被接收 到的文件為被重製的網頁字型文件。1285323 IX. Description of the invention: [Technical field to which the invention pertains] The technology of the present invention belongs to the field of computer data processing, and in particular to a computer data processing system and method for browsing Chinese files online. [Prior Art] Internet browsers are widely used to search and display electronic files. These electronic files are stored in HTML format on a computer server connected to the Internet. Computer users can access the Internet. Get these HTML format files. When a computer user accesses the Internet, they will browse some web files stored on the web server through the browser software. The user can enter a web page name (collectively referred to as a URL) to specify a website and one of the file files on it for the browser to search. When a web page file rendered in an HTML file is retrieved, the browser displays the file for the user to view. The browser soft system selects the font data in the text file by interpreting the HTML file, and then displays the text in the file according to the font data. The text displayed by the browser is formed by tracing the strokes or shapes of the fonts and fonts set in the HTML text file; if the font is not recognized by the browser, the browser will select a predetermined one. The system font is replaced, and the pre-stored system font is used to display the file. The HTML language defines a number of tags and their attributes that a web page 2 1285323 designer can use to standardize the meaning of a file sentence to render the page. However, the HTML language definition of the font is limited to those provided by the system. If the web file contains non-system fonts, the browser can only be replaced with a system-prescribed font. This alternative is described in US Patent 5,893. 915), this restriction will be ugly for displaying Chinese web pages (not (will see the original appearance) or even deformed, missing characters (some characters can not see). The Chinese characters can be represented by pictograms, and their shapes can be freely represented. For example, as shown in Figure 1, the first word ''stage'' is formed by two roots, and the two roots can be used in two types. Different fonts are used to combine the words; the second word, Bay, belongs to the engraving; the third word ''heart,' the two strokes, the peach shape, the table-show, these three words The fonts cannot be recognized by the traditional browser, so that the files that are browsed are out of shape and distorted. For the sake of the job, in the art of browsing Chinese web documents on the Internet, there is an urgent need for a system and method that can provide the original appearance of a Chinese web page. SUMMARY OF THE INVENTION The summary of the present invention is as follows: The present invention provides a method and system for a computer on the Internet to read an original file of a web page having multiple fonts. The multi-childs used by the web page provider are stored in the system or are not present in the system font. These font texts are not viewable by prior art methods and systems. In the prior art, these word systems are specified by the system. A font is substituted, and the technique of the present invention is a method for viewing the original image of a system or non-system font file 3 1285323. The system of the invention uses a webpage font server to connect to the webpage server to reconstruct the webpage file originally stored in the webpage server, so that it becomes a webpage font type HTML file, so that the webpage font browser proposed by the invention can recognize The webpage font HTML file displays the original appearance of the file. In addition to providing system font font data, the web page font server of the present invention provides non-system font font data by linking to a foreign word server via a public data network. The system arranged in this way allows the webpage file that is received by the webpage font server for the first time to find the webpage font directly from the webpage font server or indirectly finds the desired webpage font from the foreign font server. The webpage font file is formed, and the webpage font file is transmitted back to the webpage server to replace the webpage file that was originally received, so that the second received file is a reworked webpage font file.
根據本發明之一項觀點，網頁字型文件重製網頁字型 文件的HTML檔案内容係依下列步驟進行：（1)查驗文件 是否已經被建造成為網頁字型文件，若是則停止建造，否 則進行下一步驟；（2)分離及確認被接受文件裡的字型資 料；（3)根據被確認之字體字碼選用網頁字型資料來取代 原字型資料；（4)搜尋網頁字型資料，搜尋的方式有二種 情況：（一）若是搜尋系統網頁字型則從網頁字型伺服器 裡直接找尋；否則（二）轉換網頁伺服器之控制權給外接 在網路上之外字伺服器，再從耦合其上之外字資料庫中搜 尋非系統之網頁字型資料；（5)重建被找尋到之網頁字型 資料於HTML檔成為檔案内容；（6)將重製之網頁字型HTML 4 1285323 檔送回網頁伺服器。 根據本發明之另一項觀點，網頁字型HTML檔案包含 一檔頭及一檔體，檔頭記錄網頁文件之網址，文件名稱（又 稱為URL)，以及新增之網頁字型數據，其中字型數據包含 網頁字型名稱及URL位址，而檔體記錄呼應檔頭之網頁字 型名稱及URL位址，字型尺寸，文字碼，以及可供描繪文 件之文字面貌的字型資料。 根據本發明的又一項觀點，系統或非系統之字型資料 可以是屬於筆劃字型，圖樣字型，以及外圍字型。本發明 所揭露於產生筆劃字型之方法如美國專利案號6, 157,390 所述；產生圖樣字型之方法如美國專利案號6, 151，032所 述，而產生外圍字型之方法如美國專利案號6, 603, 478 B1 所述。 根據本發明的再一項觀點，網頁字型瀏覽器的軟體執 行下述功能：（1)回應上網電腦使用者的輸入指令；（2) 搜尋使用者想要之網頁字型文件；（3)解譯網頁字型HTML 檔以取得網頁字型數據；（4)產生文件之網頁字型文字； (5)顯示網頁文件。 【實施方式】 最佳實施例： 圖一示出傳統的網頁伺服器系統，其中之網頁伺服器 存放如文件32的一個例子，上網使用人之電腦係透過網 際網路22和該網頁伺服器21連接，它具有網頁瀏覽軟體 5 1285323 28來閱覽文件32，根據以下的說明，上網使用人在上述 傳統伺服器系統（先前技藝）下閱覽到的結果會如（在顯 示器23裡所示）文件36的長樣。其中顯示器23包含個 人電腦，電視機頂盒，以及其他類似的顯示儀，它們皆具 有一處理器來執行網頁瀏覽器軟體28以回應上網使用人 來閱覽一些展出在網頁伺器21上的網頁文件。如圖一所 示，網頁文件32包含三個中文字，每個字係由網頁提供 者所選或自行創建出來，文字10係由系統字碼中去挑選 然後變造其字體所獲得，字體變造的方式係取其一字根為 某一字體而另一字根為另一字體所組合而成（這種變造方 式常見於海報之製作上）；文字11係取自於印章的彫刻 體，因此它沒有系統的字碼且係屬於非系統字體；文字12 亦係選自系統，具有系統字碼唯其組成筆劃形狀用”心”的 符號取代，以突顯該文件的涵意。 根據先前技藝所述，若某文件的某文字含有不能被網 頁瀏覽器軟體認知者，則該文字的字體就屬於非系統字 體。先前技藝在處理非系統字體時，係採用一預定字體來 替代不能被網頁瀏覽器軟體認知的字體，有時網頁文件的 原貌雖被改變，然文件裡的字面意義仍然可以被閱覽到。 例如圖一所示文件32包含三個各自不同的非系統字體， 由於網頁瀏覽器軟體28僅能認知系統字碼的字以及系統 字體，所以文字10和文字12由於具有系統字碼，所以就 會被一預定的系統字體來取代它們所屬的非系統字體；而 文字10由於不具有系統字碼，所以就會被一空白字的指 6 1285323 令蓋掉而產生空的圖樣於其上，因此被閱覽到的結果就如 文件36所示，文字10和文字12由文字16和文字18所 替代，而文字11則消失在文件裡，可見先前技藝對於處 理這類網頁文件（例如32)時會完全失控，其處理結果使 得文件原貌變樣，甚至有時使得表現文件的創意完全消 失0 為了避免上述現象的發生，只有靠提供許多各式各樣 的字體，例如本發明所揭露的實施内容就是用一網頁字型 系統來建造許多被創造出來的多樣字體給網頁伺服器來 表現其上之網頁文件的文字。圖二示出本發明所提議之網 頁字型伺服器系統的組織，其中網頁字型伺服器42連結 到網頁伺服器21，用以接收網頁伺服器21裡之網頁文件， 將它們建造成網頁字型文件，再將建好之網頁字型文件傳 回網頁伺服器21裡以取代原網頁文件，當電腦使用端配 備有網頁字型瀏覽器軟體40,該瀏覽器軟體就可以解譯網 頁字型文件，並將解譯後的網頁字型文件顯示在螢幕上以 示出原貌的網頁文件。以下，茲以網頁文件32為例說明 將它建造成網頁字型文件的過程。 圖三（A)示出網頁字型伺服器42第一次從網頁伺服 器21接收到網頁文件32的HTML檔案資料，其中網頁文 件32的文字碼如圖三標示133所示係採用Big. 5碼，而 文件上的文字如標示135所示分別為A578，FA40，和4DF 的Big. 5碼，這些資料係在HTML檔頭131裡表示。這三 個文字的字體則如圖三標示153所示，分別為大字報體 7 1285323 (Post Bold)，隸書體（Lishu)，和混合符號體（Symbol Mixed)被表示在HTML檔體151裡頭。雖然HTML檔充分 表示了網頁文件32的文字内涵，但是因為大字報體，隸 書體，和混合符號體皆不屬於系統字體，所以第一次被接 收的文件經瀏覽軟體解譯後，只能認知其存在於系統之標 準字碼與字體，若文件裡含有非標準字碼，則解譯時會採 用空白的增加指令，對於非系統字體則會用一指定 (default)字體的增加指令來取代，因此圖三標示153 裡的文字就顯示如標示16和18二個文字的指定字體而標 示17由一空白字來填補。 當網頁字蜇伺服器42在第一次從網頁伺服器21接收 其上的網頁文件32後得到如圖三（A)所示之HTML檔130, 這時網頁字型伺服器42就啟動其文件檔案再製單元44(圖 四）’首先將HTML檔130裡之文字資料先行抽離，再將這 些文字資料改成網頁字型資料，文字10，11，12的網頁 字型資料分別為： 子型檔名稱·· webfontLdcw/webfont2.dcw/webfont3.dcw; 字體名稱：Post Bold/Lishu/Symbol Mixed;According to one aspect of the present invention, the HTML file content of the webpage font file remake webpage font file is performed according to the following steps: (1) checking whether the file has been constructed as a webpage font file, and if so, stopping the construction, otherwise proceeding The next step; (2) separating and confirming the font data in the accepted document; (3) selecting the web font data to replace the original font data according to the confirmed font font; (4) searching the web font data, searching There are two ways: (1) if the search system webpage font is directly searched from the webpage font server; otherwise, (2) the control of the web server is changed to the external server on the network, and then Coupling the non-systematic webpage font data in the word database; (5) reconstructing the webpage font data found in the HTML file to become the archive content; (6) recreating the webpage font HTML 4 1285323 file Send back to the web server. According to another aspect of the present invention, a webpage font HTML file includes a header and a file, a header records a webpage URL, a file name (also referred to as a URL), and a newly added webpage font data, wherein The font data includes the web page font name and the URL address, and the file body records the web page font name and URL address of the file header, the font size, the text code, and the font data of the text surface of the file to be drawn. According to still another aspect of the present invention, the system or non-system font data may be of a stroke type, a pattern font, and a peripheral font. The present invention is disclosed in U.S. Patent No. 6,157,390; the method of generating a pattern font is as described in U.S. Patent No. 6,151,032, and the method of generating peripheral fonts such as the United States. Patent No. 6, 603, 478 B1. According to still another aspect of the present invention, the software of the webpage font browser performs the following functions: (1) responding to an input command of a user of the Internet computer; (2) searching for a webpage font file desired by the user; (3) Interpreting the webpage font HTML file to obtain webpage font data; (4) generating webpage font text of the file; (5) displaying the webpage file. [Embodiment] Best Embodiment: FIG. 1 shows a conventional web server system in which a web server stores an example of a file 32, and a computer of an internet user uses an internet network 22 and the web server 21 Connected, it has a web browsing software 5 1285323 28 to view the file 32. According to the following description, the result of viewing by the Internet user under the above-mentioned conventional server system (previous skill) would be as shown in (display 23) file 36. The long sample. The display 23 includes a personal computer, a television set top box, and other similar display devices, each of which has a processor for executing the web browser software 28 in response to the Internet user to view some of the web page files displayed on the web server 21. As shown in FIG. 1, the webpage file 32 contains three Chinese characters, each of which is selected by the webpage provider or created by itself, and the text 10 is obtained by selecting and then changing the font of the system font, and the font is changed. The way is to combine the root of one character into one font and the other root into another font (this variant is often found in the production of posters); the text 11 is taken from the engraving of the seal. Therefore, it has no systematic code and belongs to non-system font; the text 12 is also selected from the system, and the system word code is replaced by the symbol of the "heart" of the stroke shape to highlight the meaning of the file. According to the prior art, if a text of a file contains a person who cannot be recognized by the web browser software, the font of the text belongs to a non-system font. In the prior art, when processing non-system fonts, a predetermined font was used to replace fonts that could not be recognized by the web browser software. Sometimes the original appearance of the web file was changed, but the literal meaning in the file could still be read. For example, the file 32 shown in FIG. 1 contains three different non-system fonts. Since the web browser software 28 can only recognize the words of the system code and the system font, the characters 10 and 12 have a system word code, so The predetermined system fonts replace the non-system fonts to which they belong; while the text 10 does not have the system word code, it will be overwritten by a blank word finger 6 1285323 to create an empty pattern on it, so it is read. The result is as shown in file 36, with text 10 and text 12 replaced by text 16 and text 18, and text 11 disappearing into the file, showing that the prior art would be completely out of control when dealing with such web page files (eg, 32), The result of the processing makes the original appearance of the file, and sometimes even makes the creative of the performance file completely disappear. 0 In order to avoid the above phenomenon, only by providing a wide variety of fonts, for example, the implementation content disclosed in the present invention uses a web page font. The system builds a number of created fonts for the web server to represent the text of the web page on it. 2 shows the organization of the proposed webpage font server system, wherein the webpage font server 42 is coupled to the webpage server 21 for receiving webpage files in the webpage server 21, and constructing them into webpage words. The type file, and then the completed webpage font file is sent back to the webpage server 21 to replace the original webpage file. When the computer end is equipped with the webpage font browser software 40, the browser software can interpret the webpage font. The file and the interpreted web page font file are displayed on the screen to show the original web page file. In the following, the webpage file 32 is taken as an example to illustrate the process of constructing it into a webpage font file. FIG. 3(A) shows that the webpage font server 42 receives the HTML file data of the webpage file 32 from the webpage server 21 for the first time, wherein the text code of the webpage file 32 is shown in FIG. The code, and the text on the file are shown as A578, FA40, and 4DF Big. 5 code, respectively, which are indicated in the HTML header 131. The fonts of these three characters are shown as 153 in Figure 3. They are the large-size bulletin 7 1285323 (Post Bold), the Lishu, and the Mixed Mixed are displayed in the HTML file 151. Although the HTML file fully expresses the textual connotation of the webpage file 32, since the large-character newspaper body, the librarian body, and the mixed symbol body are not part of the system font, the first received file can only be recognized after being interpreted by the browsing software. It exists in the standard word code and font of the system. If the file contains non-standard characters, the blank will be added when interpreting, and the non-system font will be replaced by a specified instruction of the default font. The three characters in the 153 mark are displayed as the designated fonts indicating the two characters 16 and 18, and the mark 17 is filled by a blank word. When the webpage font server 42 receives the webpage file 32 thereon from the webpage server 21 for the first time, and obtains the HTML file 130 as shown in FIG. 3(A), the webpage font server 42 starts its file archive. The re-manufacturing unit 44 (Fig. 4) first extracts the text data in the HTML file 130 first, and then changes the text data into web page type data. The web page font data of the characters 10, 11, 12 are respectively: sub-file Name·· webfontLdcw/webfont2.dcw/webfont3.dcw; Font Name: Post Bold/Lishu/Symbol Mixed;
Big5 字碼·· A578/FA4〇/A4DF，· 字型：台/灣/心， 並用這些網頁字型資料來取代被抽離的字型資料，於是得 到如圖二（B)所示新的HTML檔220，其中字體（例如字 體210)所對應字型檔（例如字型檔211)的來源及位址 也就被安置完成，因此只要具備有可解釋新HTML檔之網 8 1285323 頁瀏覽器軟體，就可重新取得這些表現網頁文件原貌的字 型資料，倘若瀏覽器軟體也有植入（plug-in)的字型產 生器（或引擎），就可讓具有原貌字型（例如圖三（B)標 示410的字型）的文件重現。 圖四示出網頁文件重製單元44的製作步驟，如方塊 51所示，網頁字型伺服器42裡有一處理器來接收第一次 從網頁伺服器21上所閱覽到的網頁文件；並有一過濾器 來抽離出網頁文件裡的文字字型；將這些被過濾出來的文 字字型依其所屬字體列表，使得每列所含文字為同一字 體，不同字體的文字分別屬於不同列。由於同一中文字可 能有許多不同字體來呈現，所以網頁文件裡若有若干不同 字體的同一中文字，則它們會分別被放置於不同列裡。在 方塊52，網頁文件重製單元44開始檢視存放在列表的每 個中文字，並對每個字型資料重製成網頁字型資料。在方 塊53,首先檢視文字的字碼及其字型的字體是否屬於已經 存在於網頁字型伺服器42之網頁字型檔案系統43裡之標 準字碼及系統字體。如果是的話，則在方塊54，直接從網 頁字型檔案系統43裡取得對應該文字字型的網頁字型來 取代，例如取代第一次被接收之網頁文件32裡的文字10 使得在方塊57，就得到如圖三（C)所示的HTML檔。如圖 三（C)標示214所示，此時之html220係仍在繼續建造中 的狀態。 如圖四方塊53所示，若被檢視的文字字碼不屬於標 準字碼以及/或者其字型字體不屬於系統字體（換言之， 9 1285323 該文字字型資料無法直接採用網頁字型檔案系統43裡網 頁字型資料來取代的狀態），則如圖二之系統架構，此時 網頁字型伺服器42會切換到另一網路服務的狀態，網頁 字型伺服器42變成公眾數據網的遠端使用者，發出請求 搜尋網頁字型資料的訊息給連結在公眾數據網上的外字 伺服器46，外字伺服器46於是根據請求訊息到其外字資 料庫47裡去搜尋想要的網頁字型資料，換言之，外字伺 服器之外字資料庫47儲存了各式各樣之非系統字型，包 括非系統的網頁字型資料，例如文字11 (圖一）的文字碼 (例如外字碼）以及其字體（隸書體）就屬於預製預存於 外字資料庫47裡的非系統字型，且如圖二所示之系統架 構圖，外字資料庫47是一直不斷地增長，若有訊息指示 某一字型字體（例如圖一的文字12)不存在於目前的外字 資料庫47裡，則外字伺服器46會透過公共數據網發出訊 息給造字中心49，將該網頁字型建造起來再傳回外字伺服 器46,再將新建造起來的網頁字型資料儲存至外字資料庫 47裡，啟動外字管理系統（如美國專利6, 603, 478 B1的 描述）來管理外字資料庫47的運作。 一旦網頁字型資料從外字資料庫47裡取得後，外字 伺服器46透過公共數據網將該網頁字型資料傳給網頁字 型伺服器42回應其請求訊息以完成外字伺服器46服務。 如圖四方塊55所示，當網頁字型伺服器42接收到從外字 伺服器46傳來的網頁字型資料後，就回復到原執行步驟 的狀態，如方塊57所示，網頁字型伺服器42裡的網頁文 1285323 · 件製單元44將非系統的網頁字型資料定義到HTML檔130 圖三（A)裡，當所有列表裡的文字皆被重製的處理後， 就會到如圖三（B)所示的HTML檔220 (稱為網頁字型基 礎的HTML檔）。 如方塊58所示，網頁字型伺服器42於是將重製完成 ‘ 之網頁字型基礎的HTML檔220傳回網頁伺服器21以取代 ， 原先的網頁文件32，使得當網頁伺服器42於第二次及其 後從網頁伺服器21接收到之網頁文件32係屬於可以原貌 • 重現之網頁字型基礎的網頁文件。 圖三（B)檔案標號211所儲存之表現網頁字形的字 型資料可以是筆劃基礎型，圖樣基礎型，或外圍形狀基礎 型。這些字型資料可藉由篏入於網頁瀏覽器軟體裡之字型 產生器（或稱為字型引擎）之解譯後產生其所對應的字 形，具有字型產生器之網頁瀏覽器軟體係存在於網頁字型 伺服器及電腦使用端，使得它們可藉由此種網頁劉覽器軟 體來解譯第二次被重製放回網頁伺服器之網頁文件的 籲 HTML檔，而可以閱覽到原貌文字字體形狀表現的中文文 件。圖五示出筆劃基礎型之一基本筆劃的例子，筆劃基礎 型之基本筆劃是由二項外顯參數··關鍵點和寬度值，以及 一項内隱參數··特徵點和曲線比例值來定義其筆劃形狀的 產生，筆劃產生的程序可用函數：y;⑷RKc；)表示，其中K 表示關鍵點，W表示寬度值，F表示特徵點，q表示曲線 比例值，如圖五標號310的方程式表示特徵點[!和F2是 由外顯參數關鍵點Κι，Κ2和寬度值Wi所決定，標號312的 11 1285323 方程式表示特徵點F3是和關鍵點K2的位置一樣的，標號 315的曲線比例值表示F 2和F 3之間的曲線形狀是由這些被 結構化起來的比例值所決定，曲線比例值是依高/低解析 度之需求所預製預存的結構化參數值，用以決定分別在高 解析度和低解析度時曲線形狀的產生，因此經由/dKFA) 的產生程序，則字型資料僅需記錄K和W，而F和Q則隱 藏於字型引擎裡用于實施的程序中，乃構成筆劃基礎型之 字型產生方法。這種筆劃基礎型的字型產生方法已經描述 於美國專利案號5, 852, 448，類似的圖樣基礎型的字型產 生方法描述於美國專利案號6, 501，475 B1，以及外圍形狀 基礎的字型產生方法描述於美國專利案號6, 151，032。 圖六所示之網頁瀏覽軟體的功能方塊圖，用來說明具 有這些功能之網頁瀏覽軟體是怎樣來解譯那些被重製後 第二次被放置在網頁伺服器上之網頁字型基礎的網頁文 件。如前述的網頁字型基礎的網頁文件係以一種新面貌的 HTML檔形式表現，此HTML檔除了保有第一次被接收之原 HTML檔的内涵外，更漆增了網頁字型的檔案及字型資料的 内容。如圖六方塊71所示，網頁字型基礎的網頁瀏覽軟 體首先回應電腦使用端的請求，這些請求包含從網頁伺服 器21選取網頁文件成為URL的一部份，於是網頁伺服器 21剖析URL的資訊後，取得及回應電腦使用端的請求，並 將被取得之第二次被放置其上的網頁字型文件傳送給網 頁瀏覽軟體40 (圖二）。在圖六方塊73，網頁字型基礎之 網頁瀏覽軟體40就對被接收到的網頁字型文件進行解 12 1285323 碼，將文件裡表達於文句之網頁字型抽離出來進行分析解 讀，如方塊74所示，網頁瀏覽軟體40有一篏入其内的字 型產生器（或稱為字型引擎），從字型資料的輸入中去產 生對應被抽離出來之網頁字型資料的字形，在方塊75將 文句依產生之字形勾劃出來，最後在方塊76裡網頁瀏覽 軟體40將整個網頁字型基礎的網頁文件依照其被描述於 網頁字型基礎之HTML檔所規範的語法及語意顯示它的文 件面貌。 一旦本發明的優先實施例已經披露且詳述如上，其他 類似或相當之想法及作法的修飾就很容易被略具本行業 技能的人士所理解，本發明的權利請求範圍不只僅限於本 實施例而已。 【圖式簡單說明】 圖一：示出先前技藝所揭露之傳統的網頁瀏覽系統。 圖二：示出本發明所揭露之網頁瀏覽系統。 圖三：（A)示出第一次從網頁伺服器接收到之HTML檔。 (B) 將它重新建造成網頁字型HTML檔的結果。 (C) 建造成網頁字型的過程。 圖四：示出揭露本發明之建造網頁字型文件的流程圖。 圖五：示出筆劃基礎字型之基本筆劃定義的範例。 圖六：示出本發明所揭露之網頁字型瀏覽軟體的流程 圖0 13 1285323 【主要元件符號說明】 21 —網頁伺服器；32—網頁文件； 10, 11，12—文件裡之文字；22—網際網路； 28—瀏覽軟體；13—電腦使用端之顯示螢幕； 36—閱覽到的文件；16, 18—閱覽到的文字； 25 —電腦使用端之輸入儀； 40—網頁字型瀏覽軟體；42—網頁字型伺服器； 43—網頁字型檔案系統；46—外字伺服器； 47-外字字型資料庫；49-字型設計中心； 51 — 58 —流程圖的不意方塊。 71 — 76 —流程圖的不意方塊。 130 —第一次收到的htm 1槽；131 —檐頭； 133 —文字字碼集；151 —槽體； 153—字體；220 —第二收到的html檔； 210—字體名稱；211 —字體的網頁字型檔； 310和312 —計算特徵點的方程式； 315 —預製預存之結構化的曲線比值。 14Big5 code·· A578/FA4〇/A4DF,· Font: Taiwan/Bay/Heart, and use these webpage fonts to replace the extracted font data, so you get the new HTML as shown in Figure 2(B) In file 220, the source and address of the font file (for example, font file 211) corresponding to the font (for example, font 210) are also set, so as long as there is a browser software that can interpret the new HTML file 8 1285323 page browser software The font data of the original web page file can be retrieved again. If the browser software also has a plug-in font generator (or engine), it can have the original font (for example, Figure 3 (B). The file of the font 410 is reproduced. 4 shows the steps of the web page file recreating unit 44. As shown in block 51, the web page font server 42 has a processor for receiving the web page file viewed from the web server 21 for the first time; The filter extracts the text fonts in the webpage file; the filtered text fonts are based on the font list of the fonts, so that the text contained in each column is the same font, and the texts of different fonts belong to different columns. Since the same Chinese character may be presented in many different fonts, if there are several different Chinese characters in the web file, they will be placed in different columns. At block 52, web page file re-creation unit 44 begins viewing each of the Chinese characters stored in the list and re-forms each font data into web page font data. At block 53, first check whether the font of the text and the font of its font belong to the standard word and system font already present in the web font file system 43 of the web page font server 42. If so, at block 54, the web page font corresponding to the text font is retrieved directly from the web font file system 43 to replace, for example, the text 10 in the web file 32 received for the first time so that at block 57 , you get the HTML file as shown in Figure 3 (C). As shown in Figure 3 (C), the html 220 system is still in the state of being built. As shown in block 53 of Figure 4, if the text code being viewed does not belong to the standard word code and/or its font type font does not belong to the system font (in other words, 9 1285323, the text type data cannot directly use the web page font file system 43 The status of the font data is replaced by the system architecture shown in Fig. 2. At this time, the webpage font server 42 switches to the state of another network service, and the webpage font server 42 becomes the remote use of the public data network. The request message for searching the webpage font data is sent to the foreign word server 46 connected to the public data network, and the foreign word server 46 then searches the foreign word database 47 according to the request message to search for the desired webpage font data, in other words. The external word server database 47 stores a variety of non-system fonts, including non-systematic web font data, such as the text code of the text 11 (Fig. 1) (such as the foreign word) and its font (library) The system is a pre-made non-system font pre-stored in the foreign word database 47, and the system architecture diagram shown in Figure 2, the external word database 47 is constantly growing. If there is a message indicating that a font font (for example, the text 12 of FIG. 1) does not exist in the current foreign word database 47, the foreign word server 46 sends a message to the word center 49 through the public data network. The font is built and sent back to the foreign word server 46, and the newly constructed web page font data is stored in the foreign word database 47, and the foreign word management system (such as the description of U.S. Patent No. 6, 603, 478 B1) is activated to manage the foreign word. The operation of the database 47. Once the web page font data is retrieved from the foreign word database 47, the foreign word server 46 transmits the web page font data to the web page font server 42 via the public data network to respond to the request message to complete the foreign word server 46 service. As shown in block 55 of FIG. 4, when the webpage font server 42 receives the webpage font data transmitted from the foreign word server 46, it returns to the state of the original execution step, as shown in block 57, the webpage font servo. Web page 1285323 in the file 42 · The component system 44 defines the non-system web page font data into the HTML file 130 in Figure 3 (A). When all the text in the list is processed by the re-processing, it will be as The HTML file 220 shown in Figure 3(B) (called the HTML file based on the web font). As shown in block 58, the web page font server 42 then forwards the re-created web page font based web file 220 back to the web server 21 to replace the original web page file 32 so that when the web server 42 is The web page file 32 received from the web server 21 twice and thereafter belongs to a web page file which can be based on the original/reproduced web page font. The font data of the web page glyph stored in the file number 211 of FIG. 3(B) may be a stroke basic type, a pattern basic type, or a peripheral shape basic type. These font data can be interpreted by a font generator (or a font engine) that is inserted into the web browser software to generate a corresponding glyph. The web browser soft system with the font generator It exists in the webpage font server and the computer use end, so that they can interpret the URL file of the webpage file that is re-posted back to the webpage server by the webpage browser software, and can be accessed. The Chinese file of the original text font shape representation. Figure 5 shows an example of one of the basic strokes of the basic stroke type. The basic stroke of the basic stroke type is composed of two explicit parameters, key points and width values, and an implicit parameter, characteristic points and curve scale values. Defining the generation of its stroke shape, the program generated by the stroke can be represented by a function: y; (4) RKc;), where K represents a key point, W represents a width value, F represents a feature point, and q represents a curve scale value, as shown in FIG. The feature points [! and F2 are determined by the explicit parameter key points Κι, Κ2 and the width value Wi, and the 11 1285323 equation of the numeral 312 indicates that the feature point F3 is the same as the position of the key point K2, and the curve scale value of the reference numeral 315 The shape of the curve between F 2 and F 3 is determined by the value of these structured values. The value of the curve is pre-stored according to the requirements of high/low resolution to determine the value of the structured parameter. The shape of the curve is generated at high resolution and low resolution, so the generation of the program via /dKFA) requires only K and W to be recorded, while F and Q are hidden in the program for implementation in the font engine. , It is the method of generating the font type of the stroke type. This type of stroke-based glyph generation method has been described in U.S. Patent No. 5,852,448, which is incorporated herein by reference to U.S. Patent No. 6, 501, 475 B1, and to The glyph generation method is described in U.S. Patent No. 6,151,032. The functional block diagram of the web browsing software shown in Figure 6 is used to illustrate how the web browsing software with these functions can interpret the webpages based on the webpage fonts that are placed on the web server for the second time after being reworked. file. The webpage file based on the aforementioned webpage font is expressed in the form of a new-faceted HTML file. In addition to retaining the connotation of the original HTML file received for the first time, the HTML file further increases the file type and file of the webpage font. The content of the type of information. As shown in block 71 of FIG. 6, the webpage font-based web browsing software first responds to the request of the computer user. The request includes selecting the webpage file from the web server 21 to be part of the URL, and the web server 21 parses the URL information. After that, the request of the computer user is obtained and responded, and the webpage font file placed on the second time is transmitted to the web browsing software 40 (Fig. 2). In block 73 of Figure 6, the webpage browsing software 40 of the webpage font type solves the 12 1285323 code of the received webpage font file, and extracts the webpage fonts expressed in the text in the document for analysis and interpretation, such as a square. 74, the web browsing software 40 has a font generator (or a font engine) inserted therein, and generates a glyph corresponding to the extracted webpage font data from the input of the font data. Block 75 delineates the sentence according to the generated zigzag. Finally, in block 76, the web browsing software 40 displays the web page based on the entire web font in accordance with the grammar and semantics of the HTML file described in the web font based. The appearance of the document. Once the preferred embodiments of the present invention have been disclosed and described in detail above, modifications of other similar or equivalent ideas and practices are readily understood by those skilled in the art, and the scope of the claims of the present invention is not limited only to the present embodiments. Only. BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a view showing a conventional web browsing system disclosed in the prior art. Figure 2: shows a web browsing system disclosed by the present invention. Figure 3: (A) shows the HTML file received from the web server for the first time. (B) Reconstructing it as the result of a web page font HTML file. (C) The process of creating a web page font. Figure 4: A flow chart showing the construction of a web page font file of the present invention. Figure 5: An example showing the basic stroke definition of the stroke basic font. Figure 6: Flow chart showing the webpage font browsing software disclosed in the present invention. 0 13 1285323 [Description of main component symbols] 21 - web server; 32 - webpage file; 10, 11, 12 - text in the file; - Internet; 28 - Browse software; 13 - Display screen of computer use; 36 - Read files; 16, 18 - Read text; 25 - Input device for computer use; 40 - Web font browsing Software; 42—Webpage font server; 43—Webpage font file system; 46—External word server; 47-External word font database; 49-character design center; 51—58 — Unexpected block of flowchart. 71 — 76 — Unexpected block of the flow chart. 130 - first received htm 1 slot; 131 - gimmick; 133 - text word code set; 151 - slot; 153 - font; 220 - second received html file; 210 - font name; 211 - font Web page fonts; 310 and 312 - Equations for calculating feature points; 315 - Pre-stored structured curve ratios. 14