EP1343132A1

EP1343132A1 - Data viewer

Info

Publication number: EP1343132A1
Application number: EP02005440A
Authority: EP
Inventors: Gary Don Thompson; Mark Johnston
Original assignee: Fujitsu Siemens Computers Inc
Current assignee: Fujitsu Siemens Computers Inc
Priority date: 2002-03-08
Filing date: 2002-03-08
Publication date: 2003-09-10

Abstract

A data viewer is provided with at least a first data processing unit. Connected to said data processing unit is a non-volatile memory unit on which at least one data file is stored. A monitor is connected to said first data processing unit for viewing data provisionally stored in a working memory. There is provided an access device in the first data processing unit which is reading a predetermined part of the data file and storing said part to the working memory so that said part is viewable on the monitor.

Description

The invention relates to a data viewer according to claim 1.
Commonly, users of data processing units are familiar with browser or data editors which download and display text from data files. Current browsers will download a complete text file and display it in the browser's main window as well as text editors will do, respectively. A scroll bar typically to the right of the text area allows the user to scroll through the file as do the scroll keys.
This approach works fine for relatively small files that contain only a few hundred or even a few thousand lines of text. The user may have to wait until the complete file will be downloaded and the user can view all of it.
However, this approach has severe limitations for extremely large files that may be hundreds of megabytes or even a few gigabytes in size. The download time will become very long. Even worse, by downloading the entire file to the client system, the disk capacity or at least the working capacity of the client system may be overwhelmed. System log files in some operating systems may grow to large sizes, for example, in Unix system the /var/adm/message files may grow to several hundred megabytes or more.
Displaying these files in a browser-based environment is thus problematic.
Similarly, the user of a data processing system has the same problem when he has to inspect a data file or text file which needs more capacity in the working memory than the system is provided with. In this case, common data or text editors load as much as possible into the working memory and write the excess to the hard disk drive.
It is therefore object of the invention to provide a data viewer which is useful in viewing large files with a minimal charge of the working memory.
This object is solved by the data viewer according to claim 1.
The data viewer according to the invention is provided with at least a first data processing unit, a non-volatile memory unit which is connected with said first data processing unit, at least one data file which is stored in said non-volatile memory unit, a monitor is connected to said first data processing unit for viewing data provisionally stored in a working memory whereby the viewer is characterised by an access device which is a part of the first data processing unit, reading a predetermined part of the data file and storing said part in the working memory unit to provide a view via said monitor.
Additional features of the invention are described as the depending claims.
In the following, the invention is further described with reference to the following drawings in which

figure 1 shows a viewer according to the invention and
figure 2 a data file viewed by said data viewer.

Figure 1 shows a client data processing unit 3 which is provided with a working memory 8, a hard disk drive 7, and a monitoring unit 5. Furthermore, in said client first data processing unit 3, an access device 9 is implemented. This access device may be implemented as a hardware as well as a software solution.
Furthermore, said client data processing device 3 is connected to a bus system 6 to which a server data processing unit 1 may be connected. In this context the system bus may be any connection between at least two data processing devices, e.g. connetions via Internet, USB, phoneline or any other communication link. The server unit 1 is additionally provided with at least a hard disk drive 2.
At least on one of the two said hard disk a data file 4 is stored for an inspection of a user of said first data processing unit 3.
In figure 2, said data file 4 is shown schematicaly. If the user wants to view a part of said data file which is determined by the boundaries 11 and 12, for instance with the scroll bar 10, he may select said part and the access device 9 will read this designated part and transfer it to the working memory 8 so that it can be shown on said monitor 5.
In a more advanced version of the invention, the access device 9 will also read a preceding part 14 and a subsequent part 15 which are adjacent parts to the designated part 13. These two parts are additionally limited by the boundary 17 for the preceding part 14 and by lower boundary 16 for the subsequent part 15.
The access device 9 reads said two additional parts and saves them to the working memory 8. This has the advantage that when the user wants to view the subsequent or preceding part, he has no waiting period until the access device has transferred one of the two parts to the working memory.
If the data viewer is provided by this additional feature after each change of the designated part to view, the access device 9 has to replace the subsequent and preceding part, respectively.
Wherever the data file is stored, the data viewer overcomes the above-mentioned problems by downloading only the designated portion of the file the user is interested to view at a time.
The user can move the scroll bar for designating the designated part 13 to any point in the file simply by adjusting the scroll bar. In the background, however, the viewer is calculating what part of said file the scroll bar's position corresponds to. It then requests just that portion of the file from the back end and displays it. It waits until the user lets it rest at the final position and then fetches that portion of the file.
Furthermore, the viewer has a search and filter capability. In a server/client system as schown in Fig. 1 schematicly, the actual searches are done on the server represented in this example by the second data processing unit 1. The client is represented by the first data processing unit 3. The search or filter results are, like the normal text, just sent to the client upon demand. Thus, the load on the working memory at the client is actually quite light. According to the invention, the viewer has the following advantages. It avoids downloading the entire file. The user doesn't have to wait for a slow download of large files and resources requirement on the client system in particular for the working memory are relatively light. It allows the user to view, search and filter the entire file without having to deal with it in blocks. Instead, to download the complete file, the access device is provided with a helper process optionally on the server system, which builds a series of indexes for the file. The indexes conceptually break the file into sequential blocks and each index points to the beginning of its respective block. The helper process then communicates information about the number of indexes to the access device which is displaying the designated part and the scroll bar. As the user moves the scroll bar, the access device calculates which index the scroll bar position corresponds to and asks the helper process to retrieve the block for the index. In this way, the access device does "just-in-time" data retrieval.
It should be mentioned that when the user moves the scroll bar, the index calculation is not done until the user lets the scroll bar rest in a position. As already said at the beginning of the description, although the description above uses a text file for the viewable object, it should be emphasized that the viewer approach would work equally well with other types of objects such as graphic images or large tables. The basic idea of having a server side helper program breaking the viewable object up into logical indexes and sending them to the client upon demand is applicable to either object types.
Finally, it should be noted that the architecture of the invention also supports text searches and filtering. The searches and filtering are done by the helper process on the server in response to a client request. The same sort of scheme with logical indexes is used to break the results of the search or filter into small chunks which are presented to the client as a seamless whole.

Claims

Data viewer with
at least a first processing unit (3),

a non-volatile memory unit (2;7) which is connected with said first data processing unit (3),

at least one data file (4) which is stored in said non-volatile memory unit (2;7),

a monitor (5) is connected to said first data processing unit (3) for viewing data, provisionally stored in a working memory (8),
characterized by
an access device (9) which is a part of said first data processing unit (3),

reading a predetermined part of the data file (4) and storing said part in the working memory (8).
Data viewer according to claim 1,
characterized in that
the access device (9) reads additionally a preceding and subsequent part of said predetermined part of said data file (4).
Data viewer according to claim 2,
characterized in that
at least a second data processing unit (1) is connected with said first data processing unit (3) and provided with said non-volatile memory unit (2;7) on which said data file (4) is stored.
Data viewer according to claim 3,
characterized in that
first data processing unit (3) is connected to the main memory unit (8) via a bus unit (6).
Data viewer according to claim 4,
characterized in that
said predetermined part of said data file (4) is determined by boundary marks viewable on said monitor unit (5).
Data viewer according to claim 5,
characterized in that
said boundary marks are represented by a scroll bar (10).