Automatic Migration Research Articles

Semantics-preserving Migration of Semantic Rules During Left Recursion Removal in Attribute Grammars

Several tools for source-to-source transformation are based on top down parsers. This restricts the user to use grammars without left recursion. Removing left recursion of a given grammar often makes it unreadable, preventing a user from concentrating on the original grammar. Additionally, the question arises, whether the tool implements the semantics of the original language, if it is implemented based on a different grammar than in the original language definition. Moreover, existing implementations of semantics for the original grammar cannot be reused directly. The paper contributes to the field of automatic migration of software (here semantic rules) induced by a grammar change. It revises removal of left recursion in the context of grammar adaptations and demonstrates, that while removing left recursion at the same time the semantic rules can be migrated automatically. Thus, a programmer can continue to use semantic rules on a left recursive grammar. The problem is explained and justified.

A study of automatic migration of programs across the Java event models

Evolution of a framework forces a change in the design of an application, which is based on the framework. The same is the case when the Java event model changed from the Inheritance model to the Event Delegation model. We summarize our experiences when attempting an automatic and elegant migration across the event models. Further, we also necessitate the need for extra documentation in patterns that will help programs evolve better.

Automatic Migration from PARMACS to MPI in Parallel Fortran Applications

The PARMACS message passing interface has been in widespread use by application projects, especially in Europe. With the new MPI standard for message passing, many projects face the problem of replacing PARMACS with MPI. An automatic translation tool has been developed which replaces all PARMACS 6.0 calls in an application program with their corresponding MPI calls. In this paper we describe the mapping of the PARMACS programming model onto MPI. We then present some implementation details of the converter tool.

Taxonomic Diversity of Pseudomonads Revealed by Computer-interpretation of Ribotyping Data

Summary In the framework of a European project entitled “High Resolution Automated Microbial Identification”, the ability of rRNA gene restriction pattern determination (ribotyping) to uncover the taxonomic diversity of a collection of bacteria was tested on 226 strains of Pseudomonas sensu lato. Improved ribotyping methodology included automated DNA extraction and purification, low voltage electrophoresis, vacuum transfer of restricted, DNA fragments to nylon membranes, hybridization with a chemically labeled probe in a hybridization incubator, computer capture of restriction patterns with automatic migration measurements, fragment size interpolation, and comparison of patterns using programs of the Taxotron package (Institut Pasteur, Paris, France). Two endonucleases were used (SmaI and HincII). The strains, belonging to more than 40 nomenspecies, displayed 169 unique rRNA gene restriction patterns with SmaI, and 159 unique rRNA gene restriction patterns with HincII. A combined analysis of both restriction data yielded 79 ribogroups or isolated ribotypes. Most (92.7%) ribogroups were homogeneous with respects to nomenspecies. Some nomenspecies were split in several ribogroups (e.g. P. putida, P. fluorescens, P. marginalis, P. pseudoalcaligenes). Nomenspecies represented by single strains gave isolated ribotypes. Biovars of P. fluorescens and P. putida formed distinct ribogroups. However, P. marginalis pathovars were not separated by ribotyping. When the ribotyping data (ribogroups) were compared to published phenotypic data (phenons) obtained with Biotype-100 strips (BioMerieux, La Balme-les-Grottes, France) and Biolog GN Microplates (Biolog, Hayward, CA), an exclusive phenon-ribogroup correspondence was observed for 13 Biotype-100 phenonslribogroups and for 14 Biolog phenonslribogroups. Other Biotype-100 or Biolog phenons showed partial correspondence with ribogroups or were split into several ribogroups. Most unclustered strains in phenotypic analyses were also unclustered in the ribotyping study. DNA-DNA hybridization was used to verify some species delineation. Ribogroup R38 which contained most of P. fluorescens biovar I strains was found to constitute a DNA hybridization group distinct from P fluorescens strains in other ribogroups or biovars. The type strain of P fluorescens, which was unclustered by ribotyping, was found isolated by DNA relatedness. Ribogroup R1 which contained most of P. putida biovar A strains was found to constitute a DNA relatedness group distinct from biovar B strains. A total of 30 isolates from polluted environment were included in the study and 9 of them were identified to six known taxons. Thirteen other isolates constituted five ribogroups not represented by a reference strain and seven isolates gave isolated ribotypes. This work showed the identification otential of ribotyping and the need to extend the ribotype database. It is now clear that the ribotype carries taxonomic information in addition to typing information.

An automatic migration scheme based on modular microcode and structured firmware sequencing

The aim of this work is to construct an automated migration system from software into firmware for functions or algorithms with complex logical structure. The approach is based on an already developed structured microcontrol scheme endowed with compound sequencing capabilities suitable for modular microprogramming. The basic idea is to extract, via compilation techniques, the sequencing structure of the function. This is expressed into complex firmware sequencing constructs such as modular calls and modular loops. Migration is then effected by sequence calls into microcode modules interpreting a target machine architecture on which the function is tested. The migration technique is implemented in a simulated environment consisting of several firmware design tools. The system has the option to produce firmware code suitable for PLA embodiment, thus facilitating function migration into VLSI microcode.

Applicability of seismic reflection sounding in regions of Precambrian geology

The paper discusses the possibilities for observing reflections from discontinuities within crystalline rocks, at depths of interest for drilling and mine-working. It is concluded that the impedance contrasts of most mafic rocks and high-density ores against felsic rocks such as granites, granodiorites and gneisses are sufficient for the generation of observable reflections. Layers are detectable if thicker than 20 m. A transition layer less than, say, 30 m thick between different rock types gives a reflection comparable to that of a sharp boundary. A major problem in interpreting the records will be the irregularity of the rock structures. It will probably be necessary to process and interpret not only reflected but also diffracted waves. The data-processing method known as automatic migration could be used to locate and delineate diffracting bodies with complex shapes. This requires the use of digital recording in the field survey. Synthetic seismograms are computed for a reflection survey over a spherical inclusion of high-density ore in granite. Waves diffracted by a single inclusion with a diameter of a few tens of meters lying at a depth of 500 m are marginally detectable at the surface, the apparent coefficient of reflection being 5–7%.

POINTE AUTOMATIQUE*

AbstractThe automatic picking of seismic sections replaces the slow and painstaking task of manual reflection plotting with rapid and economical processing by computer. Essentially, reflection picking is a decision‐making problem which also can be considered as a sort of filtering. Mathematically this means multiplying the seismic data matrix by a more or less complex decision operator.For the sake of economy, we give an initial solution for simple cases which only require easily automatized decision‐making criteria.We describe another more elaborate method for use in examining seismic phenomena which are more difficult to spot.The process used is based on the recognition of shapes and consists first of all of determining a series of characteristics capable of identifying each reflection on each trace and then of comparing these different characteristics from trace to trace so as to be able to judge the continuity or discontinuity of the reflections.Automatic picking thus leads to a schematic time‐section in which only the horizons found by calculation are retained. An automatic migration program then transforms this time‐section into a depth‐section.