ANNEX V. T 1351/04 File Search Method/FUJITSU
This decision T1351/04 – cited in the G1/19 – is in regard to “functional data” intended for controlling a technical device (in this case a computer).
The appeal in this case is against the decision of the examining division to refuse the European patent application No. 02 258 100.3.
Subject matter of the application in suit
The claimed subject matter is in regard to create a specific index-file from a csv-file (e.g. an Excel-file). By employing the index-file, it is possible to significantly increase the speed of data extraction from a large-size csv-file. In the figure 1 below, you can see an example of the csv-file (on the right hand) and of the index-file (on the left hand).
For the claimed subject matter, the csv-file requires records containing fields forming different hierarchical levels, see figure 1 above on right hand side (top hierarchical level, lower hierarchical level). The index-file is in principle built as a tree structure. This means, in the index-file information of the csv-file are connected/arranged in a tree structure. This looks schematically as follows in figure 2:
Because the records in the index-file (based on the csv-file) are arranged in the tree structure, the respective record is called a “node”. When you look in figure 1 right hand, you can see that in the csv-file there are records like “COOLING AND HEATING” and “HOUSEWORK” as the top hierarchical level. Depending on these there are further records respectively. The hierarchical records in a row of the csv-file are key character strings. The hierarchical record of the csv-file is mapped in the tree structure of the index-file. At each node there is also so-called management information, which includes information about the starting position and the number of corresponding records in the csv-file to be searched. This information permits the desired records of the csv-file to be retrieved directly when the node having the desired key character string has been found (cf para.  of the description). When only keys of high-level nodes are used it is thus not necessary to follow the tree structure all the way down to the leaf nodes in order to retrieve the desired record information.
In other words, the hierarchical structure of the csv-file is transformed in the tree structure of the index-file, wherein addition information like starting position and number of corresponding records are stored. With this arrangement information of the csv-file can be retrieved by the index-file faster.
Is the subject matter of the application technical?
Yes, since the claimed method requires the use of a computer.
Features of the subject matter contributing to the technical solution of a technical problem
To assess the inventive step it must first be considered in how far the features of the claim contribute to the solution of a technical problem. The board took a look at the following feature since it is a distinguishing feature over the prior art:
“each node in the index includes the starting position information and number information used for retrieving records”.
As mentioned above, these features are part of the index-file and are called management information.
According to the description of the present application, the invention relates to a method for “promptly searching for and extracting data from a file” (cf paragraph ). The data searched for can be of any kind, eg of a commercial nature as in the described embodiment, and thus have no technical relevance in themselves.
The data are stored as records having certain “start positions“, ie memory addresses in the file to be searched. The computer reads these addresses in the form of the “management information” in the index-file and retrieves the associated data from the csv-file to be searched. The management information thus controls the computer by directing it to a certain memory location.
Functional data, intended for controlling a technical device, are normally regarded as having technical character. The management information contained in the present claims should be regarded as contributing to the technical character of the search method according to the Board of Appeal.
It follows that the mentioned feature that has a direct bearing on how the search is conducted should be considered for inventive step.
Is the data decisive?
Further, in the decision the BoA refers to the decision T 52/85. This seems to suggest that as long as a claimed method for searching a data file is concerned with the way a computer performs the search, it may be technical. If however the kind of data is decisive, the method’s contribution is nontechnical (cf T 52/85, point 5.2). As noted above, in the present case the kind of data searched for is of no importance.
In order to support possible argumentation of the technical character of claimed subject matter comprising “functional data” it would be beneficial, when the functional data is not only disclosed in a abstract manner but also in regard to the machine-level of the computer. E.g. as mentioned above it could be disclosed that the management information controls the computer by directing it to a certain memory location.
Claim 2 of the main request in the appeal:
A computer-executable file search method for searching a file to be searched (3),
said file to be searched (3) including records having fields allocated to each of a plurality of hierarchical levels and being constructed so that records having the same key character string in a field at the same hierarchical level are arranged in series and wherein for each record, the first field is the top hierarchical level, and subsequent fields form lower hierarchical levels,
the method comprising computer-executed steps of:
- creating (S3) an index file (5) using the method of claim 1;
- accepting (S5, S6; S10, S11)) an instruction to search for data relating to a specified key character string over said file to be searched (3), the instruction including selection of either a data extraction output or a drill-down business form output;
- retrieving (S7; S12, S13) from said index file (5) management information about one or more records related to the specified key character string on said file to be searched;
- extracting (S8; S14) data of the one or more records from said file to be searched (3); and
- outputting (S9) the extracted data;
wherein the retrieving and extracting steps comprise
when the data extraction output is selected, retrieving (S7) start position information and number information as management information about records related to the specified key character string and extracting (S8) data of a number of records specified by the number information from a position specified by the start position information;
and when the drill-down business form output is selected, retrieving (S12, S13), based on the pointer, a start position of a record of the node management information of the lower hierarchical level, and extracting (S14) data of the record based on the retrieved start position of the record.