BibParser copies missing fields from previous entries

[adrian-tymorek]

I think there is a strange behavior in case some entries have missing fields. That is, suppose that entries number 1 and 4 have abstract field, but in entries 2 and 3 it's missing. In that case abstract field is added to entries 2 and 3 and it's value is copied from the first entry.
I've noticed that behavior for abstract, comment and x-color fields, but I guess it may be true for other fields too.

Here is a minimal example:

@article{ashfahani_2019_continual_DL,
	abstract = { The feasibility of deep neural networks (DNNs) to address
                  data stream problems still requires intensive study because of
                  the static and offline nature of conventional deep learning
                  approaches. A deep continual learning algorithm, namely
                  autonomous deep learning (ADL), is proposed in this paper.
                  Unlike traditional deep learning methods, ADL features a
                  flexible structure where its network structure can be
                  constructed from scratch with the absence of an initial
                  network structure via the self-constructing network structure.
                  ADL specifically addresses catastrophic forgetting by having a
                  different-depth structure which is capable of achieving a
                  trade-off between plasticity and stability. Network
                  significance (NS) formula is proposed to drive the hidden
                  nodes growing and pruning mechanism. Drift detection scenario
                  (DDS) is put forward to signal distributional changes in data
                  streams which induce the creation of a new hidden layer. The
                  maximum information compression index (MICI) method plays an
                  important role as a complexity reduction module eliminating
                  redundant layers. The efficacy of ADL is numerically validated
                  under the prequential test-then-train procedure in lifelong
                  environments using nine popular data stream problems. The
                  numerical results demonstrate that ADL consistently
                  outperforms recent continual learning methods while
                  characterizing the automatic construction of network
                  structures. },
	archiveprefix = {arXiv},
	author = {Andri Ashfahani and Mahardhika Pratama},
	comment = {published = 2018-10-17T01:40:45Z, updated = 2020-01-09T12:19:19Z},
	doi = {10.1137/1.9781611975673.75},
	eprint = {1810.07348v4},
	month = jan,
	primaryclass = {cs.LG},
	title = {Autonomous Deep Learning: Continual Learning Approach for Dynamic Environments},
	url = {http://arxiv.org/abs/1810.07348v4; http://arxiv.org/pdf/1810.07348v4},
	x-color = {#cc3300},
	x-fetchedfrom = {arXiv.org},
	year = 2019
}

@article{ashfahani_2020_DEVDAN,
	added-at = {2020-05-08T00:00:00.000+0200},
	author = {Andri Ashfahani and Mahardhika Pratama and Edwin Lughofer and Yew-Soon Ong},
	biburl = {https://www.bibsonomy.org/bibtex/2f01e837afa1ecc4df48befc53e43f458/dblp},
	ee = {https://doi.org/10.1016/j.neucom.2019.07.106},
	interhash = {d8ce7807e54d80e379324b2c3b4cd6df},
	intrahash = {f01e837afa1ecc4df48befc53e43f458},
	journal = {Neurocomputing},
	pages = {297--314},
	timestamp = {2020-05-09T11:39:11.000+0200},
	title = {DEVDAN: Deep evolving denoising autoencoder.},
	url = {http://dblp.uni-trier.de/db/journals/ijon/ijon390.html#AshfahaniPLO20},
	volume = 390,
	x-fetchedfrom = {Bibsonomy},
	year = 2020
}

[Azzaare]

Thanks! I will have a look as soon as possible.