201227587 六、發明說明： 【發明所屬之技術領域】 本發明係關於-種服務資料系統應用，特別是指利用工商 查唬貧料轉換而來的興趣點資料，並提供從原始工商 ^的資料來源進行自動異動興趣點資料的功能，透過即時 ，、政帛的智慧方法做大量資料表單異動資料的同步作業，· =種可以自動化的大量資料表單的興趣點服務工商資料同 【先前技術】 興趣點、商圈資訊傳統上需要透過人工，冤集 完整豐富而且31集耗時’也有來源正魏上的疑慮。在已習 知或是先前的技術上並無對完整的資料來源上有所發 此在對於興趣點資料來源㈣取上本篇專制提出的可謂是 特有的創新。在先前專利所提出關於資料庫異動的同步 技術上再比對異動的型態，雖然也有提出以時間變更 錄田作疋擷取異動資料的方式，來取得特定時間範圍内 動過的資料記錄，例如新增、修改或刪除的記錄，以避免： 要透過整伤貝料庫資料都重新比對或複製—次的方式達 V兩邊貝料庫。但是’當存放大量資料時如此的比對方： ，無效率的。同時，單單只有比對更新時間也不易知道= 的找刪除的資料就需要比對兩邊資料庫的所 y子大里的興趣點工商資料而言這樣是沒有效率的。 圖一為習知人X -¾丨_ > 之人工資料庫同步系統示意圖，如圖所千， 原始的資料放在來：语次士丨Α ’、 目；Ρ資料廑一 u "、貝枓庫u，其他備份的資料庫分別放在 :’ 一、目標資料庫二14、目標資料庫三15内， 服器12則有兩個方J動 責同步的機器資料庫複製伺 方法加以同步，-是定期移除目標資料庫内 201227587 資料再將整份資料庫從夾沥咨 u、目π眘粗庙 庫複製到目標資料庫- 不貝枓庫_ 14、目標資料庫三15内；或 製伺服器12必須同眸斟达 貝抖庫稷 肩问時對下游的目標資料庫一 13、目 二14'目標資料庫三15保持連線並同時 動 這樣的壞處是處理過浐沾攀❿〇义 扪更新動作。 秸的孓雜，毫無别端與後端資料庫的擴 連線問題導^ ^錯誤難以控制，若任料料錯誤發生或是 將被擴大。 不-致則後續來源與目標資料庫間的差異 圖二為習知之透過異動資料同步資料庫系統示意圖如 個改善的資料庫同步方法為，其係包含一第一 :匕以及複數個第二子系統3,其中第一子系統2係包含 ” 一二科庫2:’第二子系統3係分別包含-第二資料庫 32 -料一棟：楔組32與一資料融合模組33。資料擷取模組 係執饤-要求異動程序321 ’以便自第—子系、統2中取得 第一 > 料庫21内的資料，笼 抖第一子系統3可定時執行要求異動 =序。資枓融合模組33透過第―資料庫與第二資料庫間 貧料如第二資料311、第三資料312、第四資料犯的比對是 否存在或是否攔位有差異決定融合的方式是建立（insen)、刪 除f:)與更新(update)。此方法的缺點在於若對於擁有龐 大貝料量的貝料庫’例如本專利所針對的興趣點資料庫，兩 個貝料庫王資料跟每個欄位都做差異比對將是非常沒有效 率的負擔’同時’更新後端資料將需要跟前端來源資料做差 異比對，擁有複數個後端資料時計算量與系統負荷將會加倍 擴大；這種方法也無法達到非同時的同步資料更新，無法針 對每個後端資料庫不同的離尖峰時間做差異化的調配。 由此可見，上述使用方式仍有諸多可以改善之地方，實 非良善之設計，而亟待加以改良。 本案發明人鑑於上述方式所衍生的各項缺點，乃亟思加以改 201227587 良創新，並經多年苦心孤詣潛心研究後，決定採用目前先進 的軟體技術，以及分散式系統架構，可佇列化的中間異動檔 輸出，來研發POI業務用的工商資料系統，終於成功研發可 以有效率的自動化同步異動大量資料表單到P0I興趣指標資 料庫的工商資料系統及其方法。 h 本發明除了採用時間戳記技術來做異動資料的快速取 仔之外並、纟。合分政式儲存異動日誌、的方式，可以做到快速 區分取得新增、異動與刪除的資料項目，藉以達到更有效率 的同步一個或多個興趣點資料庫，達到自動化有效率，可進 2同時與不同時时來源工商資料與多❿肖趣點資料庫的功 能’以提供不間斷的興趣點資料中心服務。 【發明目的】 本發明之主要目的在於提供_種可自動化與有效率的 乂更新異動興趣點資料的系統及其方法，係建立—個可以 2有工商查號興趣資料彙整並轉換而得到興趣點與商圈資 枓庫的糸統’可以提供全自動化得取來源資料庫異動的作業 統，以自動從不同服務主機的刪除異動日認透過智 慧化的時間戳記方式，獲得快速、完整與正確的已刪除資料， 且可以快速與正確的得到t日異動的、 * i* ^ raL ,υ 〜市目動化同步 更新八趣點機制，可以有效且大量的得到普羅大眾有 工商資料興趣點結果。 /、的 本發明的另-目的係在於利用現有的具正確性盘_ 查號資料，轉化成興趣點資料，並且結合有效= "同步更新機制，將寶貴的工商查號資料n Ρ ^ + 體化以及數據化。 貞杆服務經驗貫 201227587 【發明内容】 為達成上述發明目之自 的系統與方法，係利 11 乂更新異動興趣點資料 n J兄有大詈盘$ ^ 興趣點來源的特性，祐处 •工商查號資料當作 更新，所以需要透過一工商資料每曰隨時都有維護 化的快速同步更# Μ ;的取彳于定期異動記錄，自動 J y旯祈興趣點資料庫， 曰刻 性，所以這是—種可以快速取得異動4持1邊資料的正確 步後端目標資料庫的服務系 二並可以自動化同 資料庫系統。其中包..、尤八適合含有龐大資料量的 能為執行記錄不同分來源刪除曰諸記錄模組，主要功 錄，包含刪除資料的時間； j除貝科的日諸記 功能為執行不同分區 Ψ主檔記錄模組，主要 筆資料-叙Μ * 〇Π ^來源歸戶檔資料寫入時，記锫下兮 華資科更新的時間並寫以^ 彳錢下该 模組，主要功能為執行在做同步異動功=刪除詢問 ，表單根據刪除時間戮記讀取里動力；=各區刪除資 “乍更新目標興趣點資料表時刪除的錚=間内的資料’ 模組’主要功能為執行在做同步異動功=科Γ異動詢問 源主要歸戶檔根據時間異動戮記向工商查號來 田 才示興趣點資料表時新增m… -異動資料輸出模組，主：新的έ己錄資料； 更新日期異動I b…仃，、.° s刪除記錄與主檔 餘士 a —更新模式騎模組，主^ 主檔異動的類 纟要功此為執行 資料庫主鍵索引θ二新/成疋更新’透過異動記錄與後端 料，若存在代表二資在貝1it示該比為新增資 要功能為執行得f丨I ' ，後端資料更新模組，主 谷同步更新目桿資#庙 過異動檔S己錄的内 本發明可以資料，完成同步作業。 動資料到興趣庫同步Γ來源是非常大量資料表單的異 趣點貝枓庫，利用來源資料庫的分散式記錄刪除 201227587 含删除時間點記錄’以及主歸戶檔表單的新增與修 間^所1職§己攔位，可以有效率的取得當日或特定時間區 點資料庫的異動記錄；透過異動記錄的取得再更新目標興趣 構你==此係以物件導向（〇bject 0riented)觀念為設計架 ^ . τ _ . 〇 °。5作為程式設計工具’使用Informix® 為興趣源資料庫’使用MiCr〇S〇ft SQL SERVER®做 料庫，搭配A貝：庫，透過0DBC方式連接兩個異質平台資 儲存新增與刀冊^日概；7多個1nformix®上北中南資料庫分區 據每签次# ，、 D,"檔，以及一個主要儲存主檔資料庫，根 記快速：量：二呆:都：記錄時間方式，即可透過該時間戮 馨Λ 综合上述整體概念以開發本系統。 式、自說，本系統結合了興趣點核心資料來源取得方 料，使以ΓΓ異動工商查號來源與目標興趣點資料庫資 的同步異動：：？且：確性的興趣點資料中心，且因為提出 得，特別、商人’疋最卽省計算成本、最有效率的異動資料取 趣點資料=具有龐Af料量的資料庫系統，大幅提升了興 ㈣身科庫的即時性與可用性。 【實施方式】 語音為杳豐富而且正確的興趣點工商資料，我們利用 來源，因ur所含的工商查號資料當作興趣點資料服務的 包二戶ίΓ;含完整又正相商家登記資訊，基本資料 商服務資料以豕名稱、地址、電話等等，額外還帶有工 等等的資訊。歹商業類別、服務項目 '產品種類、服務時間 因此，透過現有的工商查號資料來源，經由符合地理特性 201227587 的商圈景點導向的資枓整理，可以得到我們所需要的興趣點 資料庫與服務。但是，商家資料是變動的，來源的工商資料 隨時會有增減或更新’為了讓後端的興趣點資料庫也可以保 有跟來源一樣的更新以提供最新最正確的資料，同時來源的 工商資料庫含有極為龐大數量資料的特性，因此我們為了兼 顧同步更新的需求，需要找到一個最符合效率與正確性的方 法，可自動化當前端來源工商資料有異動時，可以定時的更 新後端興趣點資料庫，達到同步性，以提供最新最正 趣點資訊。 八 為了詳細說明本發明之結構以及特點所在，茲舉以下一較 佳實例’並配合相關圖示說明如後： 乂 凊先參閱圖二所示，圖三為本發明較佳實施例之自動化同 步更新異動興趣點資料系統，如圖所示’包含一個來源處理 子系統4以及至少-目標處理子系統5。其中來源處理 4係包含-來源資料庫41、多個分區日誌記錄資料庫aw 料’歸戶域更新模組4 5、分散日誌、記錄模植* 6、異動 相模.组4 7、異動記錄輸出模组4 8、刪除記錄詢問模电* 9 ·' 目標處理子系統5係包含一目標p〇I資料庫51、内 ^料川、異動資料比對模组52、異動資料讀取模組$里 動資料更新模組54。 〃 程，4係執行來源資料更新與記錄曰諸的過 =及H期來源資料庫異動資料並輸出成㈣ 有工商貧料異動發生時，歸戶主槽更新田 新到來源主檔資料庫4卜同時 會將°玄異動更 資料内：工商資料一即411代表時間標鐵到工商 t次 P 1代表该比為新增的工商資工 商貝料二即412代表該比為修改的 哈次也的商貝科。若是過期的刪 :::歸護主檔更新模組45執行刪除(―外， μ u被乂由分散曰誌記錄模組46，依據刪除資料所在 201227587 的地區分類’寫人到所在區域的日諸記錄資料庫内：例如若 是屬於北部地區工商資料，則寫入到北區曰總資料庫仏 筆育料成為北區日諸記錄421;若是屬於中部地區工商以 區曰該資㈣43’該筆資料成為中區曰訪記錄 、右疋屬於南部地區工商資料，則寫人到南區日鍵 =筆資料成為北區曰諸記錄441。此處所指稱的分區曰： 貧料庫不限定如實施例所稱三區，應用上可視 ： ::二與資料量多寡加以分區貯存，都包含在本專利運用二 ，術範圍内。同時，所存的異動日認資料421、431 = 包含該筆資料寫入的時間戳 白 利識別。 間戮。己s己錄，與異動的型態標籤，以 物4標籤符合本次同步異動時間範的料讀= 實施例中，係執行定期同步作章，目,丨思^的商貝枓。在本 商資料庫查詢工商資料當中里、4問模組47將向工 昨曰”的資料，將讀取到的前：曰異記二的曰期為’’ 錄輸出模組48。同時，定期異動作業：貝_: ’輸出到異動記 :49也會向各區曰諸記錄資料庫查詢刪：的詢問模 查詢異動類型的型態為，，刪除”的日”、動貧枓’透過 間戮記為符合本次同步異動時間範圍二工商：：該記錄時 到的刪除日誌資料421、431、44 商貝枓’將查詢 在本實施例中’係執行每曰同步迗到異動記錄模組48。 向曰諸資料庫42、43、44杳％日二、刪除詢問模組49將 删除”且刪除的時間標籤記錄的 田異動的類型為” 異動記錄輸出模組48係執行整合_ H的資料。 記錄模組49所掏取道的定期異動工商資料1模組47與刪除 组47傳入的異動資料需包含完整的工脔：中異動記錄模 d的工商資料，即原始的工商 201227587 貝料攔位或是可重建出後端p〇I =^ ^ ^ ^ 而從刪除詢問模組49來的财= 資料； 動資，即可將資料直接送給異二= :出為&期異動§己錄檔’以利後續使用者追蹤統計與報表】 統5，係 來同步 後端異動更新模組包含一個或多個目標處理子系 依據前端來源子系統4所產生的異動擋或異動資料、 更新後段目標資料庫51。 異動資料讀取模、组53係、每個後端資料庫 料讀取模組，可藉由前端輸出的異動檀案讀取：異= 接透過内部資枓傳遞方式，將前端的異動#料讀取進來一 異動資料讀取模組53讀入的異動資料屬於”刪除，，的異動； 態時，則直接交由異動資料更新模組54;若是屬於，，里 的異動型態時’則將該些資料交由異動f料比龍組Μ 判斷，當異動資料比對模組52判斷出f料是屬於，，更新 是”新增”的異動型態後’異動資料讀取模組再將這些分類 後的資料送交異動資料更新模組54作資料的合併作業。、 異動資料比對模組52係執行異動資料類型判斷的程序， 透過異動資料與目標資料的比對來判斷該筆異動資料是新辦 (inse⑴或是舊有資料的更新（update)。當異動資料，從來源^ 工商資料一 411不存在於P0I資料庫51中的p〇I資料時， 代表該筆異動資料是新增的，後續再交由異動資料更新模組 54執行新增（insert)的動作；若該筆異動是從來源的工商資料 二412 ’存在於POI資料庫51中，為p〇I資料5U，則代表 該筆異動資料是做現有資料的更新，後續再交由異動資料更 新模組54執行更新（update)的動作。 異動資料更新模組54係執行異動檔異動資料，包括新 201227587 :乍更:;==類型:對目標。〇1資料庫51進行合併的 ^ 51 § # P〇1 ::::=，_=== 源資料庫41内H 441，若是新增類型的異動時，則將來 進資料庫的動作若料―411，執行新增（insert) 41内屬於更新的工商疋二員：的異動時’則將來源資料庫 的動作。 貝’—2’執行新增（update)進資料庫 在本實施案中’根據前端異 動檔將用來同步更新後端的…= 系統’其來源皆是由前端的工商資料庫41而：：：二個後端 端依樣可接複數個目標處理统 ^的貝料，則後 轉換機制，同系j 5’透過專屬的資料過濾或 系統4與後端目標# 1端資系統資料庫。前端來源子 透過網際網路連接::料、5可在同個平台或不同平台， 資料庫亦可是#料庫41與後端目標P〇1 主 疋1J質或疋異質資料庫系統。 請參閱圖四所示，根據本專 規格的定期異動檔輸出；的異 新。後端目標處理子=Λ 步（Asynchr〇n〇us)的更 是系統負擔排定-個專屬的同:::產！:J隼時程或 =:子隨時可以安排同步“ 動擋後，貯存於_ 錄二-、異動記錄=期==錄-，、*動記 可讓監控或操作人員控制的八、时控制介面61係一 理子系統5去序列p ’工制單個或複數個目標處 ㈣中項取異動擋並更新後端目標資料庫。 11 201227587 請，圖五所示，依本發明較佳實施例之原始資料庫異動 處’其係與—來源處理子系統内部所包含 ==右包含步驟S01至步驟s〇4。在步驟=:、 枓有任何的更新需求，包括新增的卫商資料、客戶 ==二業:工_的移除，任何資料的新增修 主檔：ΐ =二。步驟S〇2中會將該筆異動更新到 中.若β刪/料*時間#錄在表單内的異動時間戳記攔位 中，右疋刪除㈣資料會交由步驟s〇3處理。在步驟如 料會先被讀出保留，刪除主黨内的該筆資料，同時交201227587 VI. Description of the invention: [Technical field to which the invention pertains] The present invention relates to the application of a service data system, in particular to information on points of interest converted from industrial and commercial materials, and provides sources of information from the original industry and commerce The function of automatic transaction point of interest data, through the instant, political and intelligent way to do a large number of data form synchronization data synchronization operations, = a variety of information forms that can be automated, point of interest service business information with [previous technology] points of interest Traditionally, business district information needs to be collected through labor, and it is full of time and 31 sets of time-consuming. There is no complete source of information on the known or prior art. This is a unique innovation in the authoritative source of interest (4). In the synchronization technique proposed in the previous patent on database change, the type of transaction is compared, although the method of changing the time to record the transaction data is also proposed to obtain the data record that has been moved within a certain time range. For example, new, modified or deleted records to avoid: To re-align or copy through the entire wounded material database - the way to the V side of the shell library. But 'when storing a lot of information, it is so much more than the other: Inefficient. At the same time, it is not efficient to find the deleted information only by comparing the update time. It is more inefficient than the information on the business points of the two points of the database. Figure 1 is a schematic diagram of the artificial database synchronization system of the traditional human X -3⁄4丨_ >, as shown in the figure, the original data is placed: the language of the gentry ', the purpose; the information 廑 a u ", shell枓 u u, other backup databases are placed in: '1, the target database 2 14, the target database 3 15, the server 12 has two sides J dynamic synchronization of the machine database replication method to synchronize , - is to periodically remove the 201227587 data in the target database and then copy the entire database from the leaching consultation, the π 慎 粗 temple library to the target database - not 枓 枓 _ 14, the target database 3 15; Or the server 12 must be connected to the downstream target database one 13 and the second 14' target database three 15 while the same is the case. ❿〇 扪 扪 update action. The noisyness of the straw, and the problem of the extension of the back-end database, is difficult to control. If any error occurs, it will be expanded. No--the difference between the follow-up source and the target database. Figure 2 is a schematic diagram of the improved database synchronization system through the transaction data system. The system includes a first: 匕 and a plurality of second children. System 3, wherein the first subsystem 2 comprises "one two libraries 2:" the second subsystem 3 includes - the second database 32 - a material: a wedge group 32 and a data fusion module 33. The module is configured to execute the request-response program 321 'to obtain the data in the first > library 21 from the first-subsystem and the system 2, and the first subsystem 3 can perform the required transaction=order. The asset integration module 33 determines whether the fusion is determined by whether the comparison between the first database and the second database, such as the second data 311, the third data 312, and the fourth data, is present or not Establish (insen), delete f:) and update (update). The disadvantage of this method is that if there is a database of interest points for a large amount of shellfish, for example, the database of points of interest for this patent, two materials It’s very ineffective to make a difference with each field. The burden of 'simultaneously' updating the back-end data will require a comparison with the front-end source data. When there are multiple back-end data, the calculation amount and system load will be doubled. This method cannot achieve non-simultaneous synchronous data update. It is impossible to make different adjustments to the different spike times of each back-end database. It can be seen that there are still many places for improvement in the above-mentioned usage methods, which are not good designs and need to be improved. The inventor of the present invention The shortcomings derived from the method are the innovations of 201227587. After years of painstaking research, we decided to adopt the current advanced software technology and decentralized system architecture, which can be used to output the intermediate shift file. The industrial and commercial information system for the development of POI business has finally successfully developed an industrial and commercial data system and method for efficiently synchronizing large-scale data forms to the P0I interest index database. h The present invention uses time stamp technology to make transaction data fast. Take aside from the squatting, squatting. It can quickly distinguish between newly added, changed and deleted data items, so as to achieve more efficient synchronization of one or more points of interest database, to achieve automation efficiency, and to enter 2 different time and time sources of industrial and commercial information. And the function of the multi-disciplinary point database to provide an uninterrupted point of interest data center service. [Object of the Invention] The main object of the present invention is to provide a system for automatically and efficiently updating the information of interest points and The method is to establish a system that can be used to collect and convert interest data from business and industry to obtain a point of interest and a business library, which can provide a fully automated source database to change the operating system to automatically change from different The service host deletes the transaction and obtains the fast, complete and correct deleted data through the intelligent time stamping method, and can quickly and correctly get the t-day change, * i* ^ raL, 市 ～ 目 目Simultaneously updating the eight interesting point mechanism can effectively and largely obtain the results of the public interest points of the public. Another object of the present invention is to convert the information into interest points by using the existing correctness disk_inspection data, and combine the effective = "synchronous update mechanism to add valuable business information to the website n Ρ ^ + Physicalization and data. Mast service experience 201227587 [Invention content] In order to achieve the above-mentioned invention system and method, the company will update the transaction point of interest data, and the characteristics of the source of interest will be The search data is regarded as an update, so it is necessary to maintain a quick synchronization through the information of a business and industry every time. It depends on the regular transaction record, and the automatic J y旯 pray point of interest database, engraving, so This is a service system that can quickly obtain the correct step back-end target database of the data of the transaction and can automatically automate the same database system. Among them, the package.., especially eight is suitable for the execution of records, different records can be deleted for the record module, the main work record, including the time to delete the data; j in addition to Beca's daily memory function for the implementation of different partitions ΨMaster file recording module, the main pen data - Μ Μ * 〇Π ^ When the source affiliation file is written, remember the time of the 兮 资 资 更新 并 并 并 并 并 并 并 并 并 并 并 并 兮 兮Execute in Synchronous Shifting = Delete Inquiries, the form reads the driving power according to the deletion time; = Delete the funds in each area. “乍 Delete the information in the target area when updating the target POI data sheet. The main function of the module is Execution in Synchronous Shifting Work = Keshu Transaction Question Source Main Ownership File According to the time shifting 向 向 向 工商 工商 才 才 才 才 才 才 才 才 才 才 才 才 才 才 才 才 才 才 才 才 才 才 才 才 才 才 才 才 才 才 才 才Recorded data; update date change I b...仃,,.° s delete record and main file shishi a-update mode ride module, main ^ main file change 的 功 为 为 执行 执行 执行 执行 执行 执行 执行New / 疋 疋 update 'transaction record The back-end material, if there is a representative of the two capital in the Bay 1it shows that the ratio of the new capital to the function is implemented f丨I ', the back-end data update module, the main valley synchronization update target rod # temple over the transaction file S recorded The present invention can be used to complete the synchronization work. The data is synchronized to the interest library. The source is a very large number of data forms. The decentralized record of the source database is used to delete the 201227587 deletion time point record and the main return. The addition and repair of the account file form can be used to efficiently obtain the transaction record of the database of the current day or the specific time zone; and obtain the target interest through the acquisition of the transaction record. Use the object-oriented (〇bject 0riented) concept as a design frame ^ τ _ . 〇 ° . 5 as a programming tool 'using Informix® for the source of interest database' using MiCr〇S〇ft SQL SERVER® library, with A Bay: Library, through the 0DBC way to connect two heterogeneous platform resources to add new and magazines ^ day; more than 7 1nformix® upper North Central South database partition according to each sign #,, D, " file, and a major Store master file Database, root record fast: quantity: two stays: all: record time mode, you can use this time to integrate the above overall concept to develop the system. The style, self-report, the system combines the core data source of interest In order to make the difference between the source of the business search and the target interest point database:: and: the point of interest of the data center, and because of the special, the merchant's most cost-effective calculation, the most Efficient transaction data for interesting point data = database system with Pang Af material volume, which greatly improves the immediacy and usability of Xing (4) body library. [Embodiment] Voice is a rich and correct point of interest industrial and commercial information, We use the source, because the industrial and commercial information contained in ur is used as the package of interest data services; including complete and positive business registration information, basic data service information, name, address, telephone number, etc. Also has information such as work and so on.歹Commercial categories, service items' product types, service hours, therefore, through the existing business search data sources, through the geographic characteristics of 201227587-oriented business district-oriented resources, we can get the database of interest points and services we need. . However, the business information is subject to change, and the source of business information will be increased or decreased or updated at any time. In order to allow the back-end point of interest database to maintain the same update as the source to provide the latest and most accurate information, and the source of the business database With the characteristics of extremely large amount of data, we need to find a method that is most in line with efficiency and correctness in order to balance the needs of synchronous update. It can automatically update the back-end point of interest database when the current source and source of business data are changed. , to achieve synchronicity, to provide the latest and most interesting information. In order to explain the structure and features of the present invention in detail, the following preferred example is described with the accompanying drawings in the following: Referring to FIG. 2, FIG. 3 is an automatic synchronization of the preferred embodiment of the present invention. The transaction point of interest data system is updated as shown in the figure 'contains a source processing subsystem 4 and at least a target processing subsystem 5. The source processing 4 system includes - source database 41, multiple partition log records database aw material 'home domain update module 4 5, scattered log, record model * 6, different phase model. group 4 7, transaction record output Module 4 8, delete record query mode power * 9 · ' target processing subsystem 5 includes a target p〇I database 51, internal material stream, transaction data comparison module 52, transaction data reading module $ The data update module 54 is activated. 〃 , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , At the same time, it will be more in the information of the mysterious movement: the industrial and commercial information one 411 represents the time standard iron to the industrial and commercial t times P 1 represents the ratio of the newly added industrial and commercial enterprises, the industrial and commercial materials, that is, 412 represents the ratio of the modified Haji also Business Bayco. If the expired deletion::: the reinstatement main file update module 45 performs the deletion ("external, μ u is deprecated by the decentralized recording module 46, according to the region classification of the deleted data where 201227587" writes the day to the region In the records database: for example, if it belongs to the industrial and commercial materials of the northern region, it will be written into the North District, the total database, and the pens will become the 421 records of the North District; if it belongs to the Central Region, the industry and commerce will use the zone (4) 43' The data becomes the record of Suwa in the Central District, and the right-hander belongs to the industrial and commercial materials in the southern region. The person who writes to the Southern District has the key to become the record of the North District. The sub-division referred to here is not limited to the example. The so-called three districts, the application of visible: :: two and the amount of data to partition storage, are included in the scope of this patent application 2, while the stored transaction data 421, 431 = contains the data written The time stamp is identified by Bailey. The 戮 戮 has been recorded, and the type label of the transaction, the object 4 label conforms to the current synchronization time of the material reading = in the embodiment, the system performs regular synchronization, the purpose,丨思^的商贝In the business database to query the industrial and commercial information, 4 questions module 47 will work for the job yesterday, the data will be read before: the different period of the second note is '' recording output module 48. , regular transaction: Baye _: 'Output to the transaction record: 49 will also query the records of each district's records database to delete: the type of the inquiry mode query transaction type is, delete the "day", the dynamic poverty Through the inter-marker, it is consistent with the scope of this synchronization transaction time. Two: The deletion log data of the record 421, 431, and 44 will be queried in the present embodiment. Module 48. To the database 42, 43, 44%, the second day, the deletion query module 49 will delete "and the deleted time tag record type of the field transaction" is the transaction record output module 48 system integration integration _ H. The record of the module 4 is the regular transaction data 1 module 47 and the deletion group 47 incoming transaction data need to contain the complete work: the business record of the transaction record d, that is, the original business 201227587 Blocking the bar or rebuilding the back end p I = ^ ^ ^ ^ and delete the financial information from the inquiry module 49; the capital, you can directly send the data to the opposite two =: the output is & the transaction § record file to facilitate subsequent user tracking Statistics and Reports] System 5, the synchronization backend update module contains one or more target processing subsystems based on the remote source or transaction data generated by the front-end source subsystem 4, and the updated target database 51. The modulo, group 53 system, and each back-end data storage module can be read by the front-end output of the transactional tamper: the difference = the internal transfer method, the front-end transaction material is read in. The transaction data read by the transaction data reading module 53 belongs to the "delete," transaction; when the state is, it is directly transferred to the transaction data update module 54; if it belongs to, the transaction type is changed. The data is judged by the transactional material, the dragon group ,, and when the transaction data comparison module 52 determines that the f material belongs to, the update is the “additional” transaction type, and the transaction data reading module then classifies the data. After the data is sent to the transaction data update module 54 for information And jobs. The transaction data comparison module 52 is a program for performing the transaction data type judgment, and the comparison between the transaction data and the target data is used to determine that the transaction data is newly created (inse (1) or old data update (update). The data, from the source ^ business information 411 does not exist in the P0I database 51 of the p〇I data, the representative of the transaction data is added, and then the subsequent transaction is updated by the transaction data update module 54 (insert) The action; if the transaction is from the source of the industrial and commercial information 2 412 'exist in the POI database 51, for the p〇I data 5U, the representative of the transaction data is to update the existing data, and then to the transaction data The update module 54 performs an update operation. The transaction data update module 54 performs the transaction data, including the new 201227587: 乍:: == type: the target 〇1 database 51 merged ^ 51 § # P〇1 ::::=, _=== H 441 in the source database 41. If there is a new type of transaction, the action of entering the database in the future will be "411", and execute (insert) 41 The two members of the industry and commerce team that are updated: Then the action of the source database. The '-2' execution is added to the database. In this embodiment, 'based on the front-end transaction file will be used to synchronize the update of the backend...=system' is sourced from the front end. Industrial and commercial database 41 and ::: The two back-ends can be connected to a plurality of target processing systems, then the conversion mechanism, the same system j 5' through dedicated data filtering or system 4 and back-end target # 1 Terminal resource database. The front-end source is connected through the Internet:: material, 5 can be on the same platform or different platforms, the database can also be #库库41 and the back-end target P〇1 main 1J quality or heterogeneous Database system. Please refer to Figure 4 for the regular transaction output according to this specification; the new target. The backend target processor = sync step (Asynchr〇n〇us) is the system load schedule - a dedicated The same as::: production!: J隼 time course or =: child can arrange synchronization at any time "after the movement, stored in _ record two -, transaction record = period == record -,, * record can be monitored or operated Personnel controlled eight, time control interface 61 is a subsystem 5 to sequence p 'work single or plural At the target (4), the middle item takes the different action and updates the backend target database. 11 201227587 Please, as shown in FIG. 5, the original database change according to the preferred embodiment of the present invention is included in the system and the source processing subsystem. == Right contains step S01 to step s〇4. In step =:, there are any update requirements, including new Guardian data, customer == two industry: work _ removal, any new data repair Main file: ΐ = 2. Step S〇2 will update the pen to the middle. If β delete / material * time # recorded in the form of the transaction time stamp block, right 疋 delete (four) data will be handed over to the step S〇3 treatment. In the steps, it will be read and retained first, and the information in the main party will be deleted.
==步驟S04決定寫入到哪區的異動曰諸中5 •錄〇 I、動資料的型態與删除的時間記錄。 請參閱圖六所示，依本發明較佳實施例之異動 包含來！資料庫定期異動檔產生相關。在步驟S11之二 始一個定期的資料庫異動讀取 —a ^ 異動，則進入步驟12、pl *疋要讀取新增修改的 在步弾S12夕士 疋哨取刪除的異動則進入步驟S13。 在乂驟S12之中，由異動記錄詢問模組依昭要 IT原=詢異動時間戳記符合的所有資料。在步驟==Step S04 determines which area of the transaction to write to. 5 • Record 〇 I, the type of the active data and the time record of the deletion. Referring to FIG. 6, the transaction according to the preferred embodiment of the present invention is included! The database is regularly related to the transaction file. In step S11 bis, a periodic database transaction reading - a ^ transaction, then proceeds to step 12, pl * 疋 to read the newly modified snippet S12 whistle whistle to delete the change, then proceeds to step S13 . In step S12, the transaction record inquiry module is in accordance with all the data that the IT original = query transaction time stamp. In the steps
艽錄〜旬I 5己錄詢問模組依照要求的異動區間跟各區曰并 d錄檔查4異動時間戳記符合 曰- 的所有資料。在步驟_ Η、^的W付合疋刪除操作 有異動型態的資料建立一定表輸出模組合併所 之Η有，貯存機制 由後二 =進行同時或非同時的同步異動更新 後端不，依本發明較佳實施例之異動檔同步更新 法係由一個或多個目標處理子系統組成 可包含步驟;料二其更新異動後端資料庫更 至乂驟S26。在步驟S21之中，則由定期異 12 201227587 =控制介面61啟動目標處理子系統進行定期同步更新異 L在步驟S22之中，藉由異動資料讀取模組依據定期 間範圍向異動檔記錄仔列讀取適合的異動記錄檔。 在乂驟S23之中，由異動資料比對模組判斷是刪除的里炉己 =是二增的異動記錄；若是新增更新的異動記錄再與 目‘ P〇I資料庫比對是否存在該異動資料決定異動㈣是新 =疋=裒在步驟S24之中，t集並合併判斷完成的異動 圮錄，右疋異質資料表單結構則進行襴位轉換 在步驟S25之中，則透過判斷完成的異：記 亲棺與目^資料表單格式對後端目標P()I資料庫進行合 報ί步驟S26之中’將處理的同步異動結果輸：完成 ，、-。果與報表，以利後續追蹤與錯誤處理。 本貫施例中的資料庫里動日々士公p 判斷…於圖三實施論動類型 紅上所述，因依本發明之興趣點資料庫同步 將:個來源處理子系統與—個或多個目標處理子'系統分成不 同部份獨立處理’故能夠在不影響後端興趣點資料庫使用之 下自動更新-個或複數個後端目標資料庫的資料，並 :二身：庫内資料的即時性與正確性；此外，本方法又提供 式的日就處理方式，可有效支援含有大量資料的資料廑 ，後端資料庫同步更新上效能的提升。另外透過中間異 方式’可提供同時或非同時的同步更新模式，可依： 料庫需求、更新時程與特性不同進衫期或是期 大大增加同f或非同ff料庫之間異動更新的可用 用性，與支援更新複數個後端目標資料庫的擴充性。 以上所述僅為符合本判_性之—實施鮮例， =制性者。任何未脫離本發明之精神與範_，輯 = 荨效修改或變更，均應包含於後附之申請專利範圍中、。 13 201227587 【特點及功效】 統及盆本方發：所提供之自動化同步更新異動興趣點資料的系 點及”方法，與其他習用技術相互比較時，更具有下列之優 丄：趣點來源資料庫是工商查號資料庫 確的來源工商資料。 人丑田，、更正 料庫與來源工商查號資料庫同步異動可保持興 趣.，“科庫景點、商圏資料的即時性與正確性。 :額習知或舊的資料庫同步更新不同與進步在於，本專 同Γΐ 有效率的得到異動資料的方法，並可以自動 同V目標的後端興趣點資料庫。 4·泫種有效率的異動檔產生方法一 =棺方式，避免若來源資料庫含有大量資:;散= 疋何種異動型態需耗費的計算與時間成本0 '' :·該有效率的異動檔產生方法 ΓΓ錄方式，可有效率的比對與得到異動:態= 與目標資料庫非具備同樣欄位構造二 連的資料^關連方式同步不同攔位結構但含有-樣序號關 7·透過刪除與智慧化判斷異動Κ 同步來源與目標資料庫。以方式’可自動化有效率的 8.後端支援的目標資料庫可以θ M 4、θ 上 列詳細說明係針對本發明 之 可行實施例之具體說 201227587 日月’惟該實施例並非用以限制本發明之專利範圍，凡未脫離 本發明技藝精神所為之等效實施或變更，均應包含 專利範圍中。 、个杀< 絲上所述，本案不但在技術思想上確屬創新，並能較 用：品增進上述多項功效，應已充分符合新穎性及進步性之 法定發明專利要件，爰依法提出申請，懇請貴局核准本件 發明專利申請案，以勵發明，至感德便。 【圖式簡單說明】 請參閱有關本發明之詳細說明及其附圖，將可進—步瞭解 本發明之技術内容及其目的功效；有關附圖為： ” 圖一為習知之人工資料庫同步系統示意圖； 圖二為習知之透過異動資料同步資料庫系統示意圖； 圖三為本發明較佳實施例之原始資料庫處理方式與 資料庫同步異動系統示意圖； 圖四本發明較佳實施例之可定期同時或非同時的同步更 新後端資料庫的系統示意圖； ’ Φ 圖五為本發明較佳實施例之來源資料庫記錄方法流程圖； 圖六為本發明較佳實施例之來源資料庫異動樓輪出方法 流程圖；以及 / 圖七為本發明較佳實施例之目標資料庫同步異動方去节 15 201227587 【主要元件符號說明】艽录〜一 I I have recorded the inquiry module in accordance with the required transaction interval and each area 曰 and d record file check 4 transaction time stamp in line with all information of 曰-. In the steps _ Η, ^ W 疋 疋 操作 操作 有 有 有 有 有 有 有 有 有 建立 建立 建立 建立 建立 建立 建立 建立 ^ ^ ^ ^ ^ , , ^ , , , , , , , , , , , , , , , , , According to a preferred embodiment of the present invention, the synchronization file synchronization update method is composed of one or more target processing subsystems, and the step of updating the transaction backend database is further step S26. In step S21, the target processing subsystem is started by the periodic interface 12 201227587 = control interface 61 to perform periodic synchronization update. In step S22, the transaction data reading module records the data according to the fixed period range. The column reads the appropriate transaction log file. In step S23, it is determined by the transaction data comparison module that the deleted memory is the second increase of the transaction record; if the newly updated transaction record is compared with the target 'P〇I database, is there a comparison? The transaction data determines the transaction (4) is new = 疋 = 裒 in step S24, t sets and merges the completed transaction record, and the right 疋 heterogeneous data form structure performs the unit conversion in step S25, then the judgment is completed. Different: record the relatives and the contents of the data form format to the back-end target P () I database to report ί in step S26 'transformed synchronization results will be processed: completed, -. Fruits and reports to facilitate follow-up and error handling. In the database of the present example, the singularity of the singularity of the singularity of the singularity of the singularity of the singularity of the singularity of the singularity of the syllabus The target processor 'system is divided into different parts and processed independently', so it can automatically update the data of one or more back-end target databases without affecting the use of the back-end point of interest database, and: two: the data in the library The immediacy and correctness; in addition, the method provides a daily processing method, which can effectively support the data containing a large amount of data, and improve the performance of the back-end database synchronization update. In addition, through the intermediate mode, it can provide simultaneous or non-simultaneous synchronous update mode, which can be updated according to the requirements of the library, the update schedule and the characteristics of the shirt or the period. The availability and support for updating the scalability of multiple back-end target repositories. The above is only for the purpose of this judgment _ sex - implementation of the fresh case, = system sex. Any changes or modifications that do not depart from the spirit and scope of the present invention are intended to be included in the scope of the appended claims. 13 201227587 [Features and Efficacy] The system and the method of automatic synchronization of the updated interest point data and the method, when compared with other conventional techniques, have the following advantages: Interest source data The library is the source of industrial and commercial information from the industrial and commercial database. The ugly field, the correction database and the source of the industrial and commercial database can be kept in sync. “The instantness and correctness of the information and business information of the library. : The difference between the amount of knowledge or the old database is different. The progress is that this is the way to get the transaction data efficiently, and it can automatically match the V target's back-end point of interest database. 4. 泫 有 有 有 有 棺 棺 棺 棺 棺 棺 棺 , , , , , , , , , , , , 来源 来源 来源 来源 来源 来源 来源 来源 来源 来源 来源 来源 来源 来源 来源 来源 来源 来源 来源 来源 来源 来源 来源 来源 来源 来源 来源 来源The different file generation method is recorded, and the efficiency can be compared and the transaction is obtained. The state = the same data with the target database is not connected with the same field structure. The related mode is synchronized with different block structures but contains the sample number. Synchronize source and target databases by deleting and intelligently judging transactions. In the manner of 'automated and efficient 8. back-end support target database can be θ M 4, θ above detailed description for the specific embodiment of the present invention 201227587 day and month 'only this embodiment is not used to limit The scope of the invention is to be construed as being limited by the scope of the invention. As mentioned above, this case is not only innovative in terms of technical thinking, but also more useful: to enhance the above-mentioned multiple functions, it should fully meet the statutory invention patent requirements of novelty and progress, and apply in accordance with the law. I urge you to approve the application for this invention patent to encourage the invention. BRIEF DESCRIPTION OF THE DRAWINGS Referring to the detailed description of the present invention and the accompanying drawings, the technical contents of the present invention and the effects of the objects of the present invention will be further understood; the related drawings are as follows: Figure 1 is a conventional artificial database synchronization FIG. 2 is a schematic diagram of a conventional system for synchronizing data through a transaction data; FIG. 3 is a schematic diagram of a method for processing a source database and a synchronization system for a database according to a preferred embodiment of the present invention; FIG. 4 is a schematic diagram of a preferred embodiment of the present invention; A schematic diagram of a system for synchronously updating the back-end database at the same time or non-simultaneously; ' Φ FIG. 5 is a flowchart of a method for recording a source database according to a preferred embodiment of the present invention; FIG. 6 is a diagram of a source database change according to a preferred embodiment of the present invention; The flow chart of the floor rotation method; and/ FIG. 7 is a synchronization data of the target database of the preferred embodiment of the present invention. 15 201227587 [Description of main component symbols]
11來源資料庫 12資料庫複製伺服器 13目標資料庫一 14目標資料庫二 15目標資料庫三 2第一子系統 21第一資料庫 211第一資料 3第二子系統 31第二資料庫 311第二資料 312第三資料 313第四資料 32資料擷取模組 321要求異動程序 33資料融合模組 4來源處理子系統 41來源資料庫 411工商資料一 412工商資料二 42北區曰誌資料庫 421北區日誌資料 43中區曰誌資料庫 431中區日誌資料 16 201227587 南區日諸資料庫 南區日t志貢料 歸戶主檔更新模組 分散日誌記錄模組 異動記錄詢問模組 異動記錄輸出模組 刪除記錄詢問模組 目標處理子系統 目標Ρ0Ι資料庫 Ρ0Ι資料 異動資料比對模組 異動資料讀取模組 異動資料更新模組 定期異動更新控制介面 異動檔序列 異動記錄一 異動記錄二 異動記錄三 1711 source database 12 database replication server 13 target database one 14 target database two 15 target database three 2 first subsystem 21 first database 211 first data 3 second subsystem 31 second database 311 Second data 312 third data 313 fourth data 32 data acquisition module 321 request transaction program 33 data fusion module 4 source processing subsystem 41 source database 411 business information one 412 business information two 42 North District Zhizhi database 421 North District Log Data 43 Central District Zhizhi Database 431 Central District Log Data 16 201227587 Southern District Daily Data Library Southern District Day tzhi tribute to the household main file update module scattered log record module transaction record query module transaction record Output module delete record query module target processing subsystem target Ρ0Ιdatabase Ρ0Ιdata transaction data comparison module transaction data reading module transaction data update module periodic transaction update control interface change file sequence change record one change record two change Record three 17